CN120299560A - A data processing method for vacuum box helium leak detection equipment - Google Patents
A data processing method for vacuum box helium leak detection equipment Download PDFInfo
- Publication number
- CN120299560A CN120299560A CN202510780001.0A CN202510780001A CN120299560A CN 120299560 A CN120299560 A CN 120299560A CN 202510780001 A CN202510780001 A CN 202510780001A CN 120299560 A CN120299560 A CN 120299560A
- Authority
- CN
- China
- Prior art keywords
- data
- segment
- points
- point
- helium
- 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
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/30—Prediction of properties of chemical compounds, compositions or mixtures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M3/00—Investigating fluid-tightness of structures
- G01M3/02—Investigating fluid-tightness of structures by using fluid or vacuum
- G01M3/04—Investigating fluid-tightness of structures by using fluid or vacuum by detecting the presence of fluid at the leakage point
- G01M3/20—Investigating fluid-tightness of structures by using fluid or vacuum by detecting the presence of fluid at the leakage point using special tracer materials, e.g. dye, fluorescent material, radioactive material
- G01M3/22—Investigating fluid-tightness of structures by using fluid or vacuum by detecting the presence of fluid at the leakage point using special tracer materials, e.g. dye, fluorescent material, radioactive material for pipes, cables or tubes; for pipe joints or seals; for valves; for welds; for containers, e.g. radiators
- G01M3/226—Investigating fluid-tightness of structures by using fluid or vacuum by detecting the presence of fluid at the leakage point using special tracer materials, e.g. dye, fluorescent material, radioactive material for pipes, cables or tubes; for pipe joints or seals; for valves; for welds; for containers, e.g. radiators for containers, e.g. radiators
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M3/00—Investigating fluid-tightness of structures
- G01M3/02—Investigating fluid-tightness of structures by using fluid or vacuum
- G01M3/26—Investigating fluid-tightness of structures by using fluid or vacuum by measuring rate of loss or gain of fluid, e.g. by pressure-responsive devices, by flow detectors
- G01M3/32—Investigating fluid-tightness of structures by using fluid or vacuum by measuring rate of loss or gain of fluid, e.g. by pressure-responsive devices, by flow detectors for containers, e.g. radiators
- G01M3/3236—Investigating fluid-tightness of structures by using fluid or vacuum by measuring rate of loss or gain of fluid, e.g. by pressure-responsive devices, by flow detectors for containers, e.g. radiators by monitoring the interior space of the containers
- G01M3/3272—Investigating fluid-tightness of structures by using fluid or vacuum by measuring rate of loss or gain of fluid, e.g. by pressure-responsive devices, by flow detectors for containers, e.g. radiators by monitoring the interior space of the containers for verifying the internal pressure of closed containers
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Theoretical Computer Science (AREA)
- Examining Or Testing Airtightness (AREA)
Abstract
The application relates to the technical field of data processing, in particular to a data processing method of vacuum box helium leak detection equipment; the method comprises the steps of obtaining helium concentration signals, filtering the helium concentration signals to obtain preprocessed concentration signals, setting reference points of data points, constructing a reference point set of the data points, calculating the credibility of the reference points in the reference point set, determining confidence weights of the reference points based on the credibility of the reference points, taking products of the confidence weights and preprocessed concentration data of corresponding reference points as local contribution values, and taking the sum of the local contribution values as a final filtering result. The application has the effect of improving the accuracy of helium concentration data processing.
Description
Technical Field
The application relates to the technical field of data processing, in particular to a data processing method of vacuum box helium leak detection equipment.
Background
In the field of precision manufacturing, it is often necessary to monitor and evaluate the sealing properties of a product. The vacuum box helium leak detection equipment is used as an efficient leak detection tool and is applied to various fields such as aerospace, automobile manufacturing and the like. The leak detection principle of the vacuum box type helium leak detection equipment is mainly based on the unique property of helium, and the helium has the characteristics of small molecular size and high diffusion speed. In the product detection process, the product is placed in a vacuum box of vacuum box helium detection equipment, a clamping cavity is formed between the product and the vacuum box, and gas in the clamping cavity is pumped out to construct a vacuum environment. And then helium is introduced into the product, under the action of air pressure, the helium can enter the vacuum box from the leakage point, and then helium concentration information in the vacuum box is collected through a sensor for detecting the helium concentration in the vacuum box, and whether the product leaks or not can be judged by analyzing the helium concentration information. Electromagnetic interference (power line, industrial equipment or wireless signals) around the equipment when collecting the helium concentration signal in the vacuum box may affect the collection of the signal by the sensor, so that noise data is generated in the helium concentration signal, and therefore denoising is usually required for the helium concentration signal before analyzing the helium concentration data.
NLM (Non-Local Means) Non-Local mean filtering algorithm is a common filtering algorithm, which can be applied to denoising of two-dimensional images and to processing of one-dimensional signals, and in the filtering process, the algorithm firstly defines a search window and a reference segment based on data points to be processed, searches segments similar to the reference segment in the window, weights and averages the segments based on similarity to update the value of the data points to be processed, and the distribution of the weights depends on the size of the similarity. The filtering result of the data points only depends on similar neighborhood data in a local range, which may cause great influence of noise data on the filtering result, especially in the case of more concentrated noise or more abnormal values, cause measurement distortion of subsequent similarity, and further may wrongly give high weight to actually uncorrelated fragments, and finally cause inaccurate filtering result.
Disclosure of Invention
In order to reduce the occurrence of inaccurate filtering in the data processing process, the application provides a data processing method of vacuum box helium leak detection equipment.
The application provides a data processing method of vacuum box helium leak detection equipment, which adopts the following technical scheme:
A data processing method of vacuum box helium leak detection equipment comprises the steps of obtaining helium concentration signals, filtering the helium concentration signals to obtain preprocessed concentration signals, setting reference points of all data points, constructing a reference point set of all data points, and calculating the credibility of the reference points in the reference point set;
The method comprises the steps of constructing a reference segment based on a reference point, constructing a neighborhood segment corresponding to the reference segment based on the position of the reference segment, calculating noise probability of the reference segment and noise probability of the neighborhood segment, determining time weight based on time intervals between data points and corresponding reference points, taking the product of the noise probability of helium concentration data in the reference segment and the noise probability of the neighborhood segment and the time weight as an abnormal degree, and processing the abnormal degree through an exponential function to obtain the reliability of the data points of the reference segment.
The method has the beneficial effects that firstly, the collected helium concentration signal is subjected to primary filtering to obtain a pretreatment concentration signal. Because the traditional filtering method is affected by noise, the filtering result is inaccurate. Therefore, the data points in the pretreatment concentration signal are readjusted to obtain the final filtering result of each data point, and the accuracy of the helium concentration data processing is improved finally.
In adjusting the data points in the pre-processed signal, one aspect is based on their neighboring reference points. In the process of obtaining the pretreatment concentration signal by primary filtering of the helium concentration signal, if more noise exists in the reference data segment corresponding to one data point, the pretreatment concentration data corresponding to the reference point is inaccurate. Therefore, in the process of adjusting the preprocessed concentration data of the data points to obtain the final filtering result, the credibility of the corresponding reference points is obtained, and the reference points are weighted based on the credibility, so that the accuracy of calculating the final filtering result is further improved.
In the calculation process of the reliability, analyzing the probability that the helium concentration data is a noise signal, and finally obtaining the reliability of the reference point based on the noise probability of the data in the reference section corresponding to the reference point.
Optionally, the step of calculating the noise probability of the helium concentration data in the reference section comprises the steps of obtaining a pressure signal in the vacuum box, calculating noise factors of data points in the helium concentration signal based on the pressure signal and the helium concentration signal, and taking the average value of the noise factors corresponding to a plurality of data points in the reference section as the noise probability of the reference section.
The method has the beneficial effects that in the process of detecting the air tightness of the product, if the product has poor tightness, leakage points exist. When helium leaks, on the one hand, a change in the concentration of helium in the vacuum box and, on the other hand, a change in the pressure in the vacuum box can be caused. In the method, the pressure signal in the vacuum box is acquired, the data points in the helium concentration signal are analyzed as noise factors of noise data by combining the change of the pressure signal and the change of the helium concentration signal, and one reference segment corresponds to a plurality of helium concentration data, so that the average value of the noise factors corresponding to the plurality of helium concentration data is taken as the noise probability of the reference segment.
Optionally, the step of calculating the noise factor of each data point in the helium concentration signal based on the pressure signal and the helium concentration signal comprises the steps of constructing a comparison data segment for any data point in the ammonia concentration signal, determining the contribution degree of the data point corresponding to the comparison data segment to the overall difference based on the overall difference by taking the standard deviation of all data in the comparison data segment as the overall difference, acquiring the data rank of the pressure data in the comparison data point in the pressure signal corresponding to the data point, taking the product of the data rank, the contribution degree and the concentration data of the pressure data at the same moment as the noise degree, and taking the normalization result of the noise degree as the noise factor.
The beneficial effect is that it usually appears as data that stands out from the surroundings for a noisy signal. Therefore, a comparison data segment corresponding to the data point is constructed based on the data point, and helium concentration data corresponding to the data point is compared with helium concentration data in the comparison data segment. The standard deviation of all the data in the comparison data segment is calculated, and represents the degree of dispersion of the data. While the value of the standard deviation is based on the value of each data, the greater the contribution of one data to the standard deviation, the more outstanding that data is in the contrast data segment. And finally, calculating the noise factor of the data point based on the contribution degree corresponding to the current data point, the pressure data and the helium concentration data.
Optionally, the step of acquiring the data ranking of the pressure data in the pressure signal corresponding to the data point in the comparison data point comprises the steps of acquiring a plurality of pressure data in the pressure signal corresponding to the comparison data segment at any moment, and ranking the rank of the pressure data corresponding to the moment as the data ranking of the pressure data at the moment based on the arrangement of the values from small to large.
The method has the beneficial effects that the data in the contrast data segment are ordered to highlight the overall level of helium concentration data of the current data point in the contrast data segment.
Optionally, the step of constructing the contrast data segment includes determining a contrast data segment lengthAcquiring a data point with the current data point as the center and the length as the lengthIs a contrast data segment.
The method has the beneficial effects that the same amount of data is selected on two sides of the data based on the current data point, so that a comparison data segment is formed.
Optionally, determining the contribution degree of the data points to the overall difference based on the overall difference includes, for any one data point, acquiring standard differences of all data except for the data points in the corresponding comparison data segment as reference differences, and taking the normalized result of the difference between the overall difference and the reference differences as the contribution degree of the data points to the overall difference.
The method has the advantages that standard deviations of all data except helium concentration data corresponding to the data points are calculated, and the contribution degree of the current data point to the standard deviation is determined by comparing the difference between the two standard deviations.
Optionally, the step of constructing the neighborhood segment corresponding to the reference segment based on the position of the reference segment includes determining a sampling time range based on the reference segment, and acquiring a preset number of data segments as the neighborhood segment within the sampling time range.
Optionally, setting reference points for each data point, and constructing reference point sets for each data point includes determining a number of reference point setsAdjacent and continuous two sides of the data pointAnd taking a set formed by a plurality of reference points as a reference point set.
The helium concentration signal optimization method has the beneficial effects that under normal conditions, the helium concentration signal is linearly changed as a whole, so that the current data point can be optimally adjusted based on adjacent data points. In the method, the data points adjacent to the two sides of the current data point are used as reference points, so that the pretreatment concentration data of the current data point is adjusted.
Optionally, the step of constructing the reference segment based on the reference point includes the step of centering a predetermined length around the reference pointAs a reference segment corresponding to a reference point.
Optionally, the step of obtaining a predetermined number of data segments within the sampling time range includes obtaining a plurality of data segments of equal length to the reference segment, calculating a mean value of the helium concentration data corresponding to the data segments, obtaining an absolute difference between the mean value and the mean value of the helium concentration data in the reference segment, and selecting a data segment having the smallest absolute differenceThe data segments are used as neighborhood segments.
The application has the following technical effects:
And performing primary treatment on the helium concentration signal to obtain a pretreatment concentration signal, and based on noise corresponding to each data point in the helium concentration signal. Therefore, the data in the preprocessed concentration signal is readjusted, so that the influence of noise on a final filtering result is reduced, and the accuracy of helium concentration signal processing is improved.
Drawings
FIG. 1 is a flow chart of a method for processing data of vacuum box helium leak detection equipment according to an embodiment of the present application.
Fig. 2 is a flow chart of a method of step S2 in a vacuum box helium leak detection apparatus data processing method.
Detailed Description
The embodiment of the application discloses a data processing method of vacuum box helium leak detection equipment, which comprises the steps of obtaining a helium concentration signal, and performing primary filtering treatment on the helium concentration signal to obtain a preprocessed concentration signal. And constructing a reference point set corresponding to each data point for each data point in the concentration signal, and carrying out secondary optimization adjustment on each data point according to the credibility of the reference point in the reference point set to finish the final filtering of the data. The reliability of the method reflects whether the primary filtering result corresponding to the data point is reliable or not, and the final filtering result can more truly reflect the potential trend of the data based on the reliability weighted filtering of the reference point.
Referring to fig. 1, a vacuum box helium leak detection apparatus data processing method includes steps S1-S3.
S1, acquiring a helium concentration signal, filtering the ammonia concentration signal to obtain a pretreatment concentration signal, setting reference points of all data points, and constructing a reference point set of all the data points.
In the process of detecting the tightness of the product, a helium concentration signal in the vacuum box is collected through a helium sensor arranged in the vacuum box. The acquisition frequency is once per second in this embodiment, and may be twice per second, three times per second, etc. in other embodiments.
And performing primary filtering treatment on the helium concentration signal by using an NLM algorithm to obtain a pretreatment concentration signal.
For any one data point, a reference point set of that data point is constructed.
For any one data point, the adjacent data point is used for optimizing the data point, so that the adjacent data point is selected as a reference point.
Determining the data amount of the reference point as. A corresponding reference data segment is determined for each data point during the initial filtering using the NLM algorithm, and the segments are weighted and averaged based on similarity by finding segments in the search box that are similar to the reference data points to update the value of the data point to be processed. The steps of the NLM algorithm are conventional technical means in the art, and are not described in detail in this embodiment. Here, theEqual to the length of the reference data segment. In the specific reference point selection process, 10 data points are selected from two sides of any data point to form a reference point. The plurality of reference points and the data points themselves constitute a set of reference points.
And S2, calculating the credibility of the reference points in the reference point set.
Referring to fig. 2, the step of calculating the credibility of the reference points in the reference point set includes steps S21-S23.
S21, constructing a reference segment based on the reference point, and constructing a neighborhood segment corresponding to the reference segment based on the position of the reference segment.
For the credibility of any one reference point, a reference segment is constructed firstly based on the reference point. Specifically, the method comprises the steps of taking a reference point as a center and taking the length as a lengthIs a reference segment. It can also be understood that the reference point is taken as the center, and both sides of the reference point are obtainedAnd data, thereby forming a reference segment corresponding to the reference point.
Based on the position of the reference point and the length of the reference segment, for any data point to be processed, two sides of the data point are provided withEach reference point corresponds to one reference segment, and each reference segment comprises a current data point (data point to be processed).
And constructing a neighborhood segment corresponding to the reference segment for the reference segment corresponding to each reference point.
And constructing a data segment adjacent to the reference segment as a neighborhood segment of the reference segment, and reflecting the credibility of the data in the reference segment through the neighborhood segment and the reference segment. Specifically, a sampling time range is determined based on the reference segment, and a preset number of data segments are acquired as neighborhood segments within the sampling time range.
In this embodiment, the sampling time range is 10 minutes before and after the sampling time of the reference point corresponding to the reference segment. Searching a plurality of data segments with equal length to the reference segment in the time period, calculating the average value of the corresponding helium concentration data in the data segments, acquiring the absolute difference value of the average value and the average value of the helium concentration data in the reference segment, and selecting the data segment with the smallest absolute difference valueThe data segments are used as neighborhood segments.
The smaller the absolute difference value is, the more similar the mean value of the helium concentration data corresponding to the reference segment and the neighborhood segment is, and the more similar the two data segments are, so that the condition of the data in the reference segment can be further reflected.
S22, calculating the noise probability of the reference segment and the noise probability of the neighborhood segment.
The step of calculating the noise probability of the helium concentration data in the reference section comprises the steps of acquiring a pressure signal in the vacuum box and calculating noise factors of data points in the helium concentration signal based on the pressure signal and the helium concentration signal.
The method comprises the steps of constructing a comparison data segment for any data point in an ammonia concentration signal, taking standard deviation of all data in the comparison data segment as an overall difference, determining contribution degree of the data point corresponding to the comparison data segment to the overall difference based on the overall difference, obtaining data ranking of pressure data in the comparison data point in a pressure signal corresponding to the data point, taking the product of the data ranking, the contribution degree and the concentration data of the pressure data at the same moment as noise degree, and taking a normalization result of the noise degree as a noise factor. Taking the average value of noise factors corresponding to a plurality of data points in the reference segment as the noise probability of the reference segment.
For any one of the data points in the helium concentration signal, the relationship between that data point and the adjacent data point is analyzed. Therefore, the contrast data segment corresponding to the data point is constructed, and in the process of constructing the contrast data segment, the length of the contrast data segment is firstly determinedAcquiring a data point with the current data point as the center and the length as the lengthIs a contrast data segment. In this embodiment, the length of the contrast data segment is 31, and 15 adjacent data points are selected on two sides of the data point, wherein the 15 adjacent data points and the data point form the contrast data segment. The relationship of the helium concentration corresponding to the data point in the contrast data segment is then analyzed.
In the comparison data segment, the standard deviation of the data in the comparison data segment represents the discrete degree of the data, and the value of each concentration data has a certain workpiece for the final standard deviation in the standard deviation calculation process. If the helium concentration data for one data point is greater, then the gain of the data in the standard deviation of the comparison data segment is greater. It is further explained that the more prominent the concentration data is in the adjacent concentration data, i.e. the concentration data of the data point is much larger or smaller than the adjacent concentration data, the greater the probability that it belongs to noise. Based on this, the contribution of the data point to the overall can reflect the likelihood that the data point is noise to some extent. Thus, the standard deviation of the overall data in the comparison data segment is calculated in this step, and the standard deviation is taken as the overall difference. Meanwhile, for a comparison data segment corresponding to one data point, calculating standard deviations of all data points except the data point, taking the standard deviations as reference differences, and taking absolute differences between the integral differences and the reference differences as contribution degrees of the data points to the integral differences.
The actual helium concentration signal itself also has some fluctuations, and when the data point is the peak value of the helium concentration signal, it is also possible to cause a larger gain on the standard deviation of the contrast data segment corresponding to the data. The pressure signal in the vacuum box is thus obtained in the present application. Based on actual scene, when the concentration of helium changes, helium gets into the vacuum box inside, and then can make the atmospheric pressure in the vacuum box change. In a real helium concentration signal, the change in helium concentration is typically accompanied by a change in pressure in the vacuum box. If one data point has a larger helium concentration, but at the same time there is no change in the corresponding pressure data, the greater the likelihood that the helium concentration at that data point will be noise. Thus in this step, the noise probability of the data point is calculated based on the contribution, helium concentration data, and pressure data within the vacuum box.
Specifically, the calculation formula of the noise probability of the data point can be expressed as:
In the formula (I), in the formula (II), A noise factor representing a data point of the helium concentration signal; Concentration data corresponding to the data points; Representing the overall difference; representing a reference difference; representing a first superparameter, primarily for preventing A zero condition occurs; representing the data ranking of the pressure data corresponding to the data points at the same time in a plurality of pressure data sequences corresponding to the comparison data segments; Representing a linear normalization function.
The data ranking refers to the order of the pressure data corresponding to the data points in the plurality of pressure data (from small to large) corresponding to the comparison data segment. For example, a set of pressure data, respectively 56.5, 55.5, 57.5, 59.5, 63.5, 62.5 and 52.5, the current data point corresponds to a value of 63.5, and the data of the plurality of data are ranked from big to small and then are ranked as 63.5, 62.5, 59.5, 57.5, 56.5, 55.5 and 52.5, and then the data rank of the current data point is named as 1.
In the formula (i) the formula (ii),The contribution degree of the concentration data corresponding to the data point to the overall difference is represented, and the larger the contribution degree is, the more discrete the concentration data corresponding to the data point is, and further the greater the possibility that the helium concentration data corresponding to the data point is noise is.
The larger the value, which represents the helium concentration to which the data point corresponds, the more prominent it is in the contrast data segment and thus the more likely it is to belong to noise data.The smaller the value, the smaller the number of bits in the sequence of the pressure data representing the same time of the data point in the sequence of the descending order of all the pressure data in the segment of the comparison data.
S23, determining time weights based on time intervals between the data points and the corresponding reference points, taking the product of the noise probability of helium concentration data in the reference section and the noise probability of the neighborhood section and the time weights as the abnormal degree, and processing the abnormal degree through an exponential function to obtain the credibility of the data points of the reference section.
One data point in the pre-processed concentration signal corresponds to a plurality of reference segments, each reference segment corresponding to one reference point. If the noise probabilities of the plurality of data in the corresponding helium concentration signal in the reference segment are all larger, the lower the accuracy of the filtering result based on the filtering of the reference segment, namely, the data value corresponding to the corresponding data point in the preprocessing concentration signal is. Meanwhile, the reliability of the data value of the reference point is also influenced by the data in the neighborhood segment corresponding to the reference segment, and if the noise probability of the data points in all the neighborhood segments corresponding to one reference segment is large, the inaccuracy of the preprocessing concentration value of the reference point is also indicated.
In addition, for a data point, if a reference segment is farther from the data point during readjustment of the preprocessed concentration data corresponding to the data point, the reference value of the reference segment is lower, so that the time interval between the data point and the reference segment should be combined simultaneously during calculation of the reliability. For this purpose, in the method, the time weight corresponding to the reference segment is calculated based on the time interval between the data point and the reference point. And finally, calculating the reliability of the final reference segment by combining the noise probability of the helium concentration corresponding to the reference segment, the noise probability of the helium concentration corresponding to the neighborhood segment and the time weight corresponding to the reference segment.
Specifically, the calculation formula of the credibility of the reference segment corresponding to one data point can be expressed as
In the formula (I), in the formula (II),Representing the corresponding first data pointThe credibility of the individual reference segments; representing data points and the first Time intervals between the reference points; Representing a second super-parameter, primarily for preventing A case of 0 occurs; Represent the first Noise probabilities of the individual reference segments; Represent the first The corresponding first reference segmentNoise probability of each neighborhood segment; Representing the number of neighborhood segments.
In the formula (i),The larger the sampling time interval between the data point and the reference point is, the lower the reference value corresponding to the corresponding reference segment is, and the lower the credibility corresponding to the reference segment is.The larger the mean value of noise probability of the helium concentration data corresponding to the reference segment is, and a large amount of noise exists in the helium concentration data corresponding to the reference segment, and the noise can influence the filtering of helium concentration signals, so that the accuracy of the preprocessed concentration data corresponding to the reference point is low. In the same way, the processing method comprises the steps of,The larger the mean value of noise probability of helium concentration data in a neighborhood segment corresponding to the reference segment is, the larger the mean value of noise probability of helium concentration data in the neighborhood segment corresponding to the reference segment is, and the lower accuracy of the corresponding reference point is further indicated.
And S3, determining confidence weights of the reference points based on the credibility of the reference points, taking the product of each confidence weight and the preprocessed concentration data of the corresponding reference point as a local contribution value, and taking the sum of a plurality of local contribution values as a final filtering result.
And for the preprocessing concentration data of one data point, adjusting through a plurality of surrounding reference points, wherein different reference points have different credibility, determining the weight corresponding to each reference point based on the credibility, and finally carrying out weighted average to obtain the final filtering result corresponding to the data point.
For the final filtering result corresponding to any one data point, the calculation formula can be expressed as follows:
In the formula (I), in the formula (II), Representing the final filtering result of the data points; Representing the corresponding first data point Confidence level of each reference point; Represent the first Preprocessing concentration data corresponding to the reference points; Representing the number of reference segments.
The reliability of the reference point is shown, and the greater the reliability, the smaller the noise probability of helium concentration data in a reference section corresponding to the reference point is, and further the higher the accuracy of pretreatment concentration data corresponding to the reference point is, and further higher weight is given.
And repeating the steps, calculating a final filtering result of each data point in the helium concentration signal, obtaining a final filtering signal, and monitoring the air tightness of the product based on the final filtering signal. Specifically, the helium concentration data may be subjected to anomaly detection using a box plot algorithm. For example, 120 data points which are continuous on a time sequence can be selected to form a box line for abnormality detection, and if 5 abnormal helium concentration data exceeding the upper limit of abnormality in the box line are detected to continuously appear on the time sequence, the leak detection equipment can be judged to be unqualified in tightness. The box map algorithm is a conventional technical means in the art, and will not be described in the present application.
The embodiment of the application also discloses a data processing method of the vacuum box helium leak detection equipment, which comprises a processor and a memory, wherein the memory stores computer program instructions, and the computer program instructions are executed by the processor to realize the data processing method of the vacuum box helium leak detection equipment.
The above system further comprises other components well known to those skilled in the art, such as a communication bus and a communication interface, the arrangement and function of which are known in the art and therefore are not described in detail herein.
The above embodiments are not intended to limit the scope of the application, so that the equivalent changes of the structure, shape and principle of the application are covered by the scope of the application.
Claims (10)
1. A data processing method of vacuum box helium leak detection equipment is characterized by comprising the steps of obtaining helium concentration signals, filtering the helium concentration signals to obtain preprocessed concentration signals, setting reference points of all data points, constructing a reference point set of all the data points, calculating the credibility of the reference points in the reference point set, determining confidence weights of all the reference points based on the credibility of the reference points, taking the product of each confidence weight and the preprocessed concentration data of the corresponding reference point as a local contribution value, and taking the sum of a plurality of local contribution values as a final filtering result;
The method comprises the steps of constructing a reference segment based on a reference point, constructing a neighborhood segment corresponding to the reference segment based on the position of the reference segment, calculating noise probability of the reference segment and noise probability of the neighborhood segment, determining time weight based on time intervals between data points and corresponding reference points, taking the product of the noise probability of helium concentration data in the reference segment and the noise probability of the neighborhood segment and the time weight as an abnormal degree, and processing the abnormal degree through an exponential function to obtain the reliability of the data points of the reference segment.
2. The method for processing helium leak detection equipment data in a vacuum tank according to claim 1, wherein the step of calculating the noise probability of the helium concentration data in the reference section comprises the steps of obtaining a pressure signal in the vacuum tank, calculating noise factors of data points in the helium concentration signal based on the pressure signal and the helium concentration signal, and taking the average value of the noise factors corresponding to a plurality of data points in the reference section as the noise probability of the reference section.
3. The method for processing the data of the vacuum box helium leak detection equipment according to claim 2, wherein the step of calculating the noise factor of each data point in the helium concentration signal based on the pressure signal and the helium concentration signal comprises the steps of constructing a comparison data segment for any data point in the ammonia concentration signal, taking standard deviations of all data in the comparison data segment as integral differences, determining contribution degrees of data points corresponding to the comparison data segment to the integral differences based on the integral differences, obtaining data ranks of pressure data in the comparison data points in the pressure signal corresponding to the data points, taking products of the data ranks, the contribution degrees and the concentration data of the pressure data at the same moment as noise degrees, and taking normalization results of the noise degrees as noise factors.
4. A method for processing data of vacuum box helium leak detection equipment according to claim 3, wherein the step of obtaining the ranking of the pressure data in the comparison data points in the pressure signals corresponding to the data points comprises the steps of obtaining a plurality of pressure data in the pressure signals corresponding to the comparison data segments at any moment, and ranking the ranking of the pressure data corresponding to the moment as the data ranking of the pressure data at the moment based on the arrangement of the values from small to large.
5. A method of processing vacuum box helium leak detection equipment data according to claim 3 wherein the step of constructing a contrast data segment comprises determining a contrast data segment lengthAcquiring a data point with the current data point as the center and the length as the lengthIs a contrast data segment.
6. A method for processing data of vacuum box helium leak detection equipment according to claim 3, wherein the step of determining the contribution degree of data points to the integral difference based on the integral difference comprises the steps of obtaining standard differences of all data except for the standard differences in corresponding comparison data segments as reference differences for any one data point, and taking the normalization result of the difference between the integral difference and the reference differences as the contribution degree of the data points to the integral difference.
7. The method for processing data of helium leak detection equipment of a vacuum box according to claim 1, wherein the step of constructing a neighborhood segment corresponding to the reference segment based on the position of the reference segment comprises the steps of determining a sampling time range based on the reference segment, and acquiring a preset number of data segments as the neighborhood segments within the sampling time range.
8. The method for processing helium leak detection equipment data of vacuum tank according to claim 1, wherein the step of setting reference points of each data point and constructing reference point sets of each data point comprises determining the number of the reference point setsAdjacent and continuous two sides of the data pointAnd taking a set formed by a plurality of reference points as a reference point set.
9. The method of claim 1, wherein the step of constructing the reference segment based on the reference point comprises the step of centering a predetermined length around the reference pointAs a reference segment corresponding to a reference point.
10. The method of claim 7, wherein the step of obtaining a predetermined number of data segments within the sampling time range comprises obtaining a plurality of data segments equal in length to the reference segment, calculating a mean value of the corresponding helium concentration data in the data segments, obtaining an absolute difference between the mean value and the mean value of the helium concentration data in the reference segment, and selecting a minimum absolute differenceThe data segments are used as neighborhood segments.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510780001.0A CN120299560B (en) | 2025-06-12 | 2025-06-12 | Data processing method for vacuum box helium leak detection equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202510780001.0A CN120299560B (en) | 2025-06-12 | 2025-06-12 | Data processing method for vacuum box helium leak detection equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN120299560A true CN120299560A (en) | 2025-07-11 |
| CN120299560B CN120299560B (en) | 2025-08-05 |
Family
ID=96283970
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202510780001.0A Active CN120299560B (en) | 2025-06-12 | 2025-06-12 | Data processing method for vacuum box helium leak detection equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN120299560B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190145929A1 (en) * | 2017-11-15 | 2019-05-16 | Infineon Technologies Ag | Sensing Systems and Methods for the Estimation of Analyte Concentration |
| CN118258787A (en) * | 2024-03-26 | 2024-06-28 | 苏州安易得安全科技有限公司 | Laser sensor data processing method based on gas concentration detection |
| CN119150206A (en) * | 2024-11-18 | 2024-12-17 | 广州泽亨实业有限公司 | Dust concentration acquisition and early warning method and system |
-
2025
- 2025-06-12 CN CN202510780001.0A patent/CN120299560B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190145929A1 (en) * | 2017-11-15 | 2019-05-16 | Infineon Technologies Ag | Sensing Systems and Methods for the Estimation of Analyte Concentration |
| CN118258787A (en) * | 2024-03-26 | 2024-06-28 | 苏州安易得安全科技有限公司 | Laser sensor data processing method based on gas concentration detection |
| CN119150206A (en) * | 2024-11-18 | 2024-12-17 | 广州泽亨实业有限公司 | Dust concentration acquisition and early warning method and system |
Non-Patent Citations (2)
| Title |
|---|
| 张炜, 邢玉安, 邢如英: "地下水中氡含量离散度变化与地震的短临前兆", 地震学报, no. 03, 15 June 1987 (1987-06-15) * |
| 王晓峰;张国毅;王然;: "一种新的未知雷达信号快速分选方法", 电子信息对抗技术, no. 05, 15 September 2011 (2011-09-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN120299560B (en) | 2025-08-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117272210A (en) | A data detection method and system for abnormal hidden dangers in building construction | |
| Saragiotis et al. | Automatic P phase picking using maximum kurtosis and/spl kappa/-statistics criteria | |
| CN116413395B (en) | An intelligent detection method for environmental anomalies | |
| CN118134539B (en) | User behavior prediction method based on intelligent kitchen multi-source data fusion | |
| CN116304963B (en) | Data processing system suitable for geological disaster early warning | |
| CN116432144B (en) | Data fusion system based on carrier communication | |
| CN111858990B (en) | Wafer map failure mode similar retrieval method based on convolution classification network | |
| CN120668082B (en) | A method to reduce the monitoring error of uneven settlement in deep foundation pits | |
| CN117648657B (en) | A method for optimizing multi-source data processing in urban planning | |
| CN117573667A (en) | A method and system for optimizing and cleaning data between coal mines | |
| CN120299560B (en) | Data processing method for vacuum box helium leak detection equipment | |
| CN117556274A (en) | A method for analyzing temperature data anomalies of heat pipe backplanes | |
| CN117541626A (en) | Remote sensing image change detection method based on deep learning | |
| CN117031566A (en) | A method and system for constructing geomagnetic anomaly data sets | |
| CN116471660A (en) | WiFi Fingerprint Indoor Localization Method Based on Denoising Autoencoder and Convolutional Neural Network | |
| CN116402825A (en) | Bearing fault infrared diagnosis method, system, electronic equipment and storage medium | |
| CN118194151A (en) | Fault diagnosis method for aircraft engine bearings based on graph multi-scale approximate entropy | |
| CN110426695B (en) | Target classification method based on radar identification database characteristic curve fitting | |
| CN117540008A (en) | Contract abnormal data risk intelligent analysis method | |
| CN121365260B (en) | A method and system for monitoring the assembly process of a sunroof production line based on digital twins. | |
| CN120910728B (en) | Intelligent analysis method for operation data of numerical control machine tool | |
| CN119295822B (en) | A method for monitoring the molten pool during the SLM melting process | |
| CN116992389B (en) | A method and system for detecting false data in the Internet of Things | |
| CN118277943B (en) | Rapid detection method for motor running state | |
| CN121770174A (en) | Electric energy data real-time monitoring system of factory power distribution cabinet |
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 |