CN117330882A - Automatic test method and system for filter - Google Patents
Automatic test method and system for filter Download PDFInfo
- Publication number
- CN117330882A CN117330882A CN202311631339.7A CN202311631339A CN117330882A CN 117330882 A CN117330882 A CN 117330882A CN 202311631339 A CN202311631339 A CN 202311631339A CN 117330882 A CN117330882 A CN 117330882A
- Authority
- CN
- China
- Prior art keywords
- waveform
- evaluation
- output
- time
- curve
- 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
- 238000010998 test method Methods 0.000 title claims abstract description 17
- 238000011156 evaluation Methods 0.000 claims abstract description 195
- 238000012360 testing method Methods 0.000 claims abstract description 66
- 238000000034 method Methods 0.000 claims abstract description 23
- 230000000877 morphologic effect Effects 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 claims description 2
- 238000005259 measurement Methods 0.000 abstract description 3
- 230000002159 abnormal effect Effects 0.000 description 24
- 238000004458 analytical method Methods 0.000 description 9
- 238000009826 distribution Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 230000002776 aggregation Effects 0.000 description 3
- 238000004220 aggregation Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/003—Environmental or reliability tests
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R23/00—Arrangements for measuring frequencies; Arrangements for analysing frequency spectra
- G01R23/16—Spectrum analysis; Fourier analysis
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Environmental & Geological Engineering (AREA)
- Mathematical Physics (AREA)
- Investigating Strength Of Materials By Application Of Mechanical Stress (AREA)
Abstract
The invention relates to the technical field of filter measurement and test, in particular to an automatic test method and system for a filter. The method analyzes the difference of output waveform curves corresponding to different filters and the difference of the output waveform curves and standard waveform curves to obtain waveform moment evaluation at different moments; all the waveform curves are overlapped to obtain an overlapped waveform curve, and different waveform segments to be detected are obtained by intercepting the overlapped waveform curve according to the fluctuation evaluation range; analyzing morphological differences between any two output waveform curve sections corresponding to the waveform section to be tested to obtain discrete evaluation between the two output waveform curve sections; and determining the overall evaluation of the waveform segment to be measured according to the discrete evaluation among all the output waveform curve segments corresponding to the waveform segment to be measured. The invention improves the accuracy of the automatic test of the filter.
Description
Technical Field
The invention relates to the technical field of filter measurement and test, in particular to an automatic test method and system for a filter.
Background
The automatic test of the filter is a method for evaluating the performance and reliability of the filter, and the test and analysis are performed by utilizing computer control and automation technology, so that the problem that test evaluation is inaccurate due to subjective factors of test evaluation caused by different experiences of testers can be avoided.
The automatic test generally involves analyzing and evaluating test data to obtain performance parameters and indexes of the filter, but a large amount of data generated by a large amount of test is directly analyzed one by one, so that the change condition of the test parameters generated by the large amount of test is difficult to embody, and the abnormal condition of the whole data is required to be judged through the parameters of the large amount of test, so that the purpose of obtaining more comprehensive filter performance evaluation is achieved.
By automated testing, a fast, accurate and repeatable assessment of filter performance can be achieved. Which can reduce human error and test time and provide more comprehensive test coverage. Meanwhile, the automatic test can provide more comprehensive support for filter performance evaluation and improvement strategies through batch test and large-scale data analysis.
The current common method for automatically testing the filter is to identify abnormal test parameters of the current batch of filter samples through outlier variation differences among a large number of test parameters, but when analyzing the batch of test parameters, the abnormal parameters of part of samples are doped in a large number of normal parameters, and the problem that the difference of partial abnormal values is insufficient for detection through a fixed threshold exists.
Disclosure of Invention
In order to solve the technical problem that when abnormal test parameters of a current batch of filter samples are identified through outlier variation differences among a large number of test parameters, abnormal parameters of part of samples are doped in a large number of normal parameters, so that the difference of part of abnormal values is insufficient to detect through a fixed threshold, the invention aims to provide an automatic test method and an automatic test system for the filter, and the adopted technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides an automated test method for a filter, the method comprising the steps of:
obtaining an output waveform curve of the filter output;
at different moments, analyzing the differences of the output waveform curves corresponding to different filters and the differences of the output waveform curves and the standard waveform curves to obtain waveform moment evaluations at different moments;
sequencing waveform time evaluations at different times to obtain a descending sequence; determining a fluctuation evaluation range according to the difference of the evaluation of adjacent two waveform moments in the descending sequence; all the waveform curves are overlapped to obtain an overlapped waveform curve, and different waveform segments to be detected are obtained by intercepting the overlapped waveform curve according to the fluctuation evaluation range;
analyzing morphological differences between any two output waveform curve sections corresponding to the waveform section to be tested to obtain discrete evaluation between the two output waveform curve sections; and determining the overall evaluation of the waveform segment to be measured according to the discrete evaluation among all the output waveform curve segments corresponding to the waveform segment to be measured.
Preferably, the calculation formula of the waveform moment evaluation is as follows:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Waveform time evaluation of the output waveform curve at time t 2; />Is an S-shaped function; />The frequency corresponding to the ith output waveform curve at the time t 2; />The frequency corresponding to the output waveform curve at the time t2 is obtained; />The maximum value of the frequency corresponding to the time t2 of all the output waveform curves is obtained; />Corresponding to all output waveform curves at the time t2A minimum value of the frequency of (2); />The frequency value of the standard waveform curve a at the time t 2; />The frequency value of the standard waveform curve a at the previous moment of the moment t 2; />Standard deviation of all output frequencies in the time t 2; />The number of output waveform curves for all filter outputs.
Preferably, the sorting the waveform time evaluations at different time to obtain a descending sequence includes:
and (5) carrying out descending order arrangement on waveform time evaluations at different time to obtain a descending order sequence.
Preferably, the determining the fluctuation evaluation range according to the difference between the adjacent two waveform time evaluations in the descending order includes:
calculating the difference value of waveform moment evaluation of two adjacent positions based on the descending sequence, and marking the difference value as an evaluation difference value; recording the waveform moment evaluation with the minimum corresponding sequence in the two waveform moment evaluations corresponding to the maximum evaluation difference as a target evaluation;
and determining a fluctuation evaluation range according to the target evaluation and the maximum waveform moment evaluation.
Preferably, the determining the fluctuation evaluation range according to the target evaluation and the maximum waveform time evaluation includes:
taking the position corresponding to the maximum waveform moment evaluation in the descending sequence as one end point of the fluctuation evaluation range; and taking the position corresponding to the target evaluation as the other end point of the fluctuation evaluation range, and obtaining the fluctuation evaluation range according to the two end points.
Preferably, the step of obtaining different waveform segments to be measured by cutting in the superimposed waveform curve according to the fluctuation evaluation range includes:
acquiring a plurality of waveform moment evaluations in a fluctuation evaluation range, and recording the waveform moment evaluations as moment evaluations to be selected; evaluating and calculating a middle range value for two adjacent time points to be selected; and intercepting a plurality of different waveform segments to be detected by taking the time sequences corresponding to the two adjacent intermediate range values as endpoints.
Preferably, the analyzing the morphological difference between any two output waveform curve segments corresponding to the waveform segment to be tested to obtain a discrete evaluation between the two output waveform curve segments includes:
obtaining the minimum matching distance of two output waveform curve segments by a dynamic time warping algorithm on any two output waveform curve segments; obtaining the matching distance of each position in the two output waveform curve sections, and recording the matching distance larger than the minimum matching distance as a larger distance;
the calculation formula of the discrete evaluation is as follows:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>For the output waveform curve section->Discrete evaluation of (2); n is the number of the waveform curve segments which are output every two and correspond to the waveform segments to be detected; />For a larger distance->Is the number of (3);n time points t contained in the current waveform segment X to be detected; />Is a larger distance; />Is formed by two output waveform curve segmentsIs the minimum matching distance of (2); />The position of the time evaluation to be selected contained in the current waveform segment X to be detected is screened out from the fluctuation evaluation range; />Matching distance for each position of current two output waveform curve sections>Is the maximum position of (2); d is a Euclidean distance function.
Preferably, the determining the overall evaluation of the waveform segment to be measured according to the discrete evaluation between all the output waveform curve segments corresponding to the waveform segment to be measured includes:
and taking the average value of the discrete evaluation among all the output waveform curve sections corresponding to the waveform section to be measured as the integral evaluation of the waveform section to be measured.
Preferably, the determining the overall evaluation of the waveform segment to be measured according to the discrete evaluation between all the output waveform curve segments corresponding to the waveform segment to be measured further includes:
if only one output waveform curve section is left, performing discrete evaluation calculation on the output waveform curve section and a corresponding standard waveform curve, and taking the obtained result value as the overall evaluation of the waveform section to be measured.
In a second aspect, an embodiment of the present invention provides an automated test system for a filter, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor implementing an automated test method for a filter as described above when executing the computer program.
The embodiment of the invention has at least the following beneficial effects:
firstly, obtaining an output waveform curve output by a filter; under different moments, analyzing the differences of output waveform curves corresponding to different filters and the differences of the output waveform curves and standard waveform curves to obtain waveform moment evaluation at different moments, wherein the waveform moment evaluation reflects the frequency influence generated in local time under the condition that outliers are generated at a single moment; sequencing waveform time evaluations at different times to obtain a descending sequence; according to the difference of the evaluation of adjacent two waveform moments in the descending sequence, determining a fluctuation evaluation range, and screening out a range with overlarge output frequency difference under a single moment; intercepting different waveform segments to be detected in the superimposed waveform curve according to the fluctuation evaluation range; analyzing morphological differences between any two output waveform curve sections corresponding to the waveform section to be tested to obtain discrete evaluation between the two output waveform curve sections, wherein the discrete evaluation reflects the abnormal necessity of a filter test performance sample; and determining the overall evaluation of the waveform segment to be measured according to the discrete evaluation among all the output waveform curve segments corresponding to the waveform segment to be measured. According to the invention, the resolution analysis of different fluctuation difference paragraphs is performed through the discrete fluctuation conditions of the data generated by aggregation among a plurality of samples, so that the visual analysis is performed according to the fluctuation condition difference significance in each section, and the accuracy of the filter automatic test is improved.
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 flow chart of a method for an automated test method for filters according to one embodiment of the present invention;
fig. 2 is a schematic diagram of a test apparatus according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purposes, the following description refers to the specific implementation, structure, characteristics and effects of an automatic testing method and system for a filter according to the present invention with reference to 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 embodiment of the invention provides an automatic test method and a specific implementation method of a system for a filter. The invention aims to solve the technical problem that when the abnormal test parameters of the current batch of filter samples are identified through the outlier variation difference among a large number of test parameters, the abnormal parameters of part of the samples are doped in a large number of normal parameters, so that the difference of part of abnormal values is insufficient to detect through a fixed threshold value. According to the invention, the resolution analysis of different fluctuation difference paragraphs is performed through the discrete fluctuation conditions of the data generated by aggregation among a plurality of samples, so that the visual analysis is performed according to the fluctuation condition difference significance in each section, and the accuracy of the filter automatic test is improved.
The following specifically describes a specific scheme of an automatic testing method and system for a filter provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of steps of an automatic testing method for a filter according to an embodiment of the present invention is shown, the method includes the steps of:
step S100, an output waveform curve of the filter output is obtained.
Firstly, installing test equipment to obtain test data of a current batch of samples.
The filter test is performed by selecting proper test equipment and instruments, including equipment such as a signal generator, an oscilloscope, a spectrum analyzer and the like, and related sensors, cables and the like.
For filter test, a priori unified test waveform is adopted as an input waveform of the filter, external test environments such as temperature, humidity and the like in a laboratory in the test process are unified, when the external test environments and the input waveform are unified, the obtained filter test parameters only have output waveform differences generated by abnormal filter performance, abnormal data analysis in the filter test is more accurate after fusion, and abnormal data erroneous judgment caused by external environment interference is reduced.
Referring to fig. 2, fig. 2 is a schematic diagram of a test apparatus, a filter is fixed on the test apparatus, a in fig. 2 is a signal input end, B is a signal output end, C is a filter sample, D is a fixture, the fixture is for fixing the filter sample to be tested, a priori unified test waveform is input to the signal input end, the signal output end is connected with an oscilloscope, and is used for inspecting an output waveform of the filter, and the output waveform is also called an output waveform curve. And further, an output waveform curve of the filter output is obtained.
And step 200, analyzing the difference of the output waveform curves corresponding to the different filters and the difference of the output waveform curves and the standard waveform curves at different moments to obtain waveform moment evaluation at different moments.
The waveforms output by the same input signal processing by the same batch of filters should be the same, which means that the processing capacity of each filter is the same. However, in the actual filtering process, product quality differences exist among all test samples, so that abnormal fluctuation is identified by taking the data fluctuation condition of part of normal samples as a reference, and further, the abnormal samples are screened according to the local differences of output waveform matching.
According to the invention, abnormal parameters are accurately identified for the fluctuation tolerance characteristics through the observation data with different dimensions, and the reliability self-adaptive evaluation of the system state is carried out according to the distribution of the abnormal parameters.
The case where there is a difference between the output waveform curves reflects the difference of the test filters for the same input waveform processing, and thus the splitting of the case of different fluctuations is performed by the case of fluctuation difference between the output waveforms of the filters. Since the input waveform to be tested is fixed and the design parameters of the filter are also fixed, a priori standard output waveform curve, abbreviated as standard waveform curve, can be derived from the input waveform and the design parameters, denoted as a. Taking the moment when the filter receives the filter output as the output starting moment t1 of the filter sample; the time at which the test input signal is input to the current filter is taken as the input start time t1' of the present sample.
Judging the time delay condition of the sample: since the difference between the input starting time t1' represents the input time of the input signals of different filters, the difference t1-t1' between the output starting time t1 and the input starting time t1' reflects the output time delay difference between the filters, and after the time delay difference is obtained, the difference between the output waveforms is specifically required to be judged, so that the test parameters of the filters are more comprehensively analyzed.
And taking the output starting time t1 of each sample output waveform curve as the starting time, and superposing all the output waveform curves to obtain a superposed waveform curve. Analyzing a superimposed waveform curve distributed in a sample space: the curve frequency difference appears at the same moment indicates that partial abnormal frequency exists to generate outlier data so as to influence the frequency change in a certain time range, so that the abnormal condition at a single moment can be judged through the frequency difference at the single moment, and further the frequency influence generated by the condition of generating outlier at the single moment in the local time sequence is analyzed.
For each output waveform curve, taking the output starting moment as a reference point, calculating waveform moment evaluation at the moment t2 from any moment t2, specifically, analyzing differences of the output waveform curves corresponding to different filters and differences of the output waveform curves and standard waveform curves at different moments to obtain waveform moment evaluation at different moments.
The method comprises the steps of carrying out a first treatment on the surface of the Wherein,is an S-shaped function; />The frequency corresponding to the output waveform curve at the time t2 is obtained; />The frequency corresponding to the ith output waveform curve at the time t 2; />The maximum value of the frequency corresponding to the time t2 of all the output waveform curves is obtained;the minimum value of the frequency corresponding to the time t2 of all the output waveform curves is obtained; />The frequency value of the standard waveform curve a at the time t 2; />The frequency value of the standard waveform curve a at the previous moment of the moment t 2; />Standard deviation of all output frequencies in the time t 2; />The number of output waveform curves output for all filters, i.e., the number of filter samples tested.
Wherein,the absolute value of the frequency difference is weighted between the current time t2 and the previous time t2-1 for the maximum and minimum frequency difference output at the current time t 2. Although it is difficult to achieve the same for each filter sample at the time of production and there is no error with the a priori waveform, the small differences between samples reflect similar output frequencies at the same time. When the transfusion is in the same timeWhen the difference between the maximum output frequencies is 0, reflecting that the output frequencies of the filter samples at the current moment are equal, and judging that no output fluctuation exists at the moment; similarly, when there is no difference between the standard output frequencies at the front and rear time, it is not necessary to determine the performance difference of the filter.
For the deviation characteristic of the current output frequency around the standard output frequency generated by the difference of the current time test sample relative to the standard, the difference is enlarged by the power of 4, the sign difference indicates that the amplitude of the numerical value is larger than that of the standard, and the numerical value is larger to indicate that more numerical values larger than the standard exist.
The superimposed output waveforms are started from the next moment with the starting moment as a reference, the fluctuation evaluation of each moment is calculated, the distribution similarity of the fluctuation evaluation is analyzed, and the waveform moment evaluation of different moments is obtained.
Step S300, sorting waveform time evaluations at different time to obtain a descending sequence; determining a fluctuation evaluation range according to the difference of the evaluation of adjacent two waveform moments in the descending sequence; and superposing all the waveform curves to obtain a superposed waveform curve, and intercepting different waveform segments to be detected in the superposed waveform curve according to the fluctuation evaluation range.
Firstly, sorting waveform time evaluations at different time to obtain a descending sequence, specifically: and (5) carrying out descending order arrangement on waveform time evaluations at different time to obtain a descending order sequence.
Since the position with the high waveform time evaluation is determined as the time with the excessive frequency difference at the single time, the waveform time evaluation corresponding to all the time is sorted in descending order, and the descending order sequence is obtained.
Further, according to the difference of the evaluation of the adjacent two waveform moments in the descending order, the fluctuation evaluation range is determined, and the method is specific:
calculating the difference value of waveform moment evaluation of two adjacent positions based on the descending sequence, and marking the difference value as an evaluation difference value; will be the maximum evaluation differenceThe waveform time evaluation with the minimum corresponding sequence in the two corresponding waveform time evaluation is recorded as a target evaluation; and determining a fluctuation evaluation range according to the target evaluation and the maximum waveform moment evaluation. Specific: calculating the fluctuation evaluation difference value of two adjacent positions,wherein->For evaluation at the j-th waveform instant in the descending sequence; />For evaluation at the j+1th waveform moment in the descending order; />Evaluating the difference value of two adjacent waveform moments in the descending sequence; acquisition->Will beThe waveform time evaluation with the smaller serial number of the two corresponding waveform time evaluations is marked as +.>。
According to the difference of the evaluation of adjacent two waveform moments in the descending sequence, determining a fluctuation evaluation range, and specifically: taking the position corresponding to the maximum waveform moment evaluation in the descending sequence as one end point of the fluctuation evaluation range; and taking the position corresponding to the target evaluation as the other end point of the fluctuation evaluation range, and obtaining the fluctuation evaluation range GP according to the two end points.
Thus, the fluctuation evaluation range GP is obtained. And intercepting different waveform segments to be detected in the superimposed waveform curve according to the fluctuation evaluation range. And intercepting each position in the fluctuation evaluation range GP on the time sequence of the superimposed output waveform curve according to the time sequence intermediate range value of each two adjacent GP positions to obtain a plurality of waveform segments to be measured. Namely, acquiring a plurality of waveform moment evaluations in a fluctuation evaluation range, and recording the waveform moment evaluations as moment evaluations to be selected; evaluating and calculating a middle range value for two adjacent time points to be selected; and intercepting a plurality of different waveform segments to be detected by taking the time sequences corresponding to the two adjacent intermediate range values as endpoints. It should be noted that, the calculation of the intermediate range value is a well-known technique for those skilled in the art, and will not be described herein. It should be noted that, the superimposed output waveform curve is the superimposed waveform curve, and specific steps are already given in step S200, and will not be described herein.
So far, the difference between the waveform moment evaluations is analyzed, and then the overlapped output waveforms are intercepted, so that a plurality of waveform segments to be detected for identification are obtained.
Step S400, analyzing morphological differences between any two output waveform curve segments corresponding to the waveform segments to be tested to obtain discrete evaluation between the two output waveform curve segments; and determining the overall evaluation of the waveform segment to be measured according to the discrete evaluation among all the output waveform curve segments corresponding to the waveform segment to be measured.
And analyzing different waveform segments to be tested, wherein the more compact the distribution of the frequency values contained in the waveform segments to be tested, the more similar the fluctuation characteristic, and the more obvious the difference between the waveform segments to be tested.
First from the beginning of the waveform segment to be measuredBeginning to the end of the waveform segment to be measured +.>In the superimposed waveform curve of the corresponding waveform segment to be measured, the process of arbitrarily carrying out pairwise comparison reflects the condition of abnormal filter performance generated in the waveform segment to be measured.
And analyzing any two output waveform curve segments corresponding to the waveform segments to be tested, and comparing deformation conditions existing between the curves to serve as positions of abnormal filter processing performance. It should be noted that the output waveform curve corresponding to the waveform segment to be measured is an output waveform curve segment which is consistent with the end point of the waveform segment to be measured.
Arbitrarily selecting two different output waveform curve segmentsComparing, namely obtaining a comparison result of the two output waveform curve segments by a dynamic time warping algorithm (DTW), wherein the comparison result is the minimum matching distance ∈10 of the two output waveform curve segments>。
Analyzing the information obtained by matching in the waveform segment to be tested, wherein the matching distance of each position in the two output waveform curve segments in the part of the waveform segment to be testedThe smaller the relative minimum matching distance L, the fewer the deviation feature formed at that location.
First selecting matching distanceDistance +.>Recorded as greater distance +.>. A larger distance is selected at +.>More, the frequency of the abnormality generated by the current filter sample in the waveform section to be detected is higher, and the fluctuation evaluation of the moment corresponding to the larger distance in the current partial deformation section and the position +_ contained in the waveform section to be detected are shown>The more similar the corresponding fluctuation evaluation is, the more the discrete condition generated by the data fluctuation in the waveform segment to be tested is close to the data point distribution discrete condition of the larger position in the waveform segment to be tested, and the difference of the waveform segment to be tested as the filter test performance sample is provedThe higher the necessity of the paragraph. And a distance less than or equal to L represents a situation where the matching error is smaller, which is a more trusted situation.
And analyzing the morphological difference between any two waveform segments to be measured to obtain discrete evaluation between the two waveform segments to be measured.
The calculation formula of the discrete evaluation is as follows:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>For the output waveform curve section->Discrete evaluation of (2); n is the number of the waveform curve segments which are output every two and correspond to the waveform segments to be detected; />For a larger distance->Is the number of (3);n time points t contained in the current waveform segment X to be detected; />Is a larger distance; />Is formed by two output waveform curve segmentsIs the minimum matching distance of (2); />The position of the time evaluation to be selected contained in the current waveform segment X to be detected is screened out from the fluctuation evaluation range; />Matching distance for each position of current two output waveform curve sections>Is the maximum position of (2); d is a Euclidean distance function. It should be noted that, since the two ends of each to-be-measured waveform segment are intermediate range values, and each to-be-measured waveform segment corresponds to one to-be-selected time evaluation, the to-be-selected time evaluation is also the waveform time evaluation within the fluctuation evaluation range, so that the position of one to-be-selected time evaluation can be obtained from the current to-be-measured waveform segment.
Wherein,for the ratio of the number of positions with larger distance contained in the current waveform segment to the number of all positions in the current waveform segment to be measured, the larger the ratio is, the larger deviation situation is generated due to the fact that the whole numerical value in the current waveform segment to be measured has excessive numerical values, the more abnormal situations are generated in the waveform segment to be measured, and the more filters in the output waveform curve segment represented in the waveform segment to be measured are reflected to generate work abnormality.
Wherein,the difference between a plurality of larger distances and the minimum matching distance in the waveform segment to be detected is reflected, the difference represents the discrete difference condition of the output frequency, so that the larger the calculated difference accumulation is, the more obvious the data discrete condition in the segment is reflected. />For the time distance between the position of the waveform time evaluation to be selected with higher P in the current waveform time evaluation to be tested and the maximum matching distance position in the waveform segment to be tested, when the distance is smaller, the positions of the fluctuation anomalies in the segment are similar, and the probability of the occurrence of the anomalies is higher in the data discrete condition representation in the waveform segment to be tested.
After the discrete evaluation between the two output waveform curve segments is obtained, a segment with high discrete evaluation needs to be selected from the discrete evaluation segments to serve as an abnormal data segment of the current batch filter test sample to be output.
And (3) optionally analyzing two output waveform curve segments from the current waveform segment to be tested to obtain discrete evaluation. If only one output waveform curve segment remains, the output waveform curve segment is compared with the standard waveform curve corresponding to the waveform segment to be measured. And taking the average value of the discrete evaluation among all the output waveform curve segments corresponding to the waveform segment to be measured as the discrete evaluation of the waveform segment to be measured.
And (3) inputting all the discrete evaluation into a normalization function for normalization, and outputting normalization results between [0,1 ].
After the normalized discrete evaluation is obtained, different normalized output results can be endowed with different colors through the existing visual distribution technology so as to facilitate the subsequent analysis. Visualization elements are added to further enhance the expressive power of the analysis results. To show overall trends, etc.
In summary, the present invention relates to the technical field of filter measurement and test. Firstly, obtaining an output waveform curve output by a filter; at different moments, analyzing the differences of the output waveform curves corresponding to different filters and the differences of the output waveform curves and the standard waveform curves to obtain waveform moment evaluations at different moments; sequencing waveform time evaluations at different times to obtain a descending sequence; determining a fluctuation evaluation range according to the difference of the evaluation of adjacent two waveform moments in the descending sequence; all the waveform curves are overlapped to obtain an overlapped waveform curve, and different waveform segments to be detected are obtained by intercepting the overlapped waveform curve according to the fluctuation evaluation range; analyzing morphological differences between any two output waveform curve sections corresponding to the waveform section to be tested to obtain discrete evaluation between the two output waveform curve sections; and determining the overall evaluation of the waveform segment to be measured according to the discrete evaluation among all the output waveform curve segments corresponding to the waveform segment to be measured. According to the invention, the resolution analysis of different fluctuation difference paragraphs is performed through the discrete fluctuation conditions of the data generated by aggregation among a plurality of samples, so that the visual analysis is performed according to the fluctuation condition difference significance in each section, and the accuracy of the filter automatic test is improved.
The embodiment of the invention also provides an automatic test system for the filter, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the steps of the method when executing the computer program. Since a detailed description of an automatic test method for a filter is given above, a detailed description is omitted.
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 (9)
1. An automated test method for a filter, the method comprising the steps of:
obtaining an output waveform curve of the filter output;
at different moments, analyzing the differences of the output waveform curves corresponding to different filters and the differences of the output waveform curves and the standard waveform curves to obtain waveform moment evaluations at different moments;
sequencing waveform time evaluations at different times to obtain a descending sequence; determining a fluctuation evaluation range according to the difference of the evaluation of adjacent two waveform moments in the descending sequence; all the waveform curves are overlapped to obtain an overlapped waveform curve, and different waveform segments to be detected are obtained by intercepting the overlapped waveform curve according to the fluctuation evaluation range;
analyzing morphological differences between any two output waveform curve sections corresponding to the waveform section to be tested to obtain discrete evaluation between the two output waveform curve sections; determining the overall evaluation of the waveform segment to be measured according to the discrete evaluation among all the output waveform curve segments corresponding to the waveform segment to be measured;
the calculation formula of the waveform moment evaluation is as follows:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Waveform time evaluation of the output waveform curve at time t 2; />Is an S-shaped function; />The frequency corresponding to the ith output waveform curve at the time t 2; />The frequency corresponding to the output waveform curve at the time t2 is obtained; />The maximum value of the frequency corresponding to the time t2 of all the output waveform curves is obtained; />The minimum value of the frequency corresponding to the time t2 of all the output waveform curves is obtained; />The frequency value of the standard waveform curve a at the time t 2; />The frequency value of the standard waveform curve a at the previous moment of the moment t 2; />Standard deviation of all output frequencies in the time t 2; />The number of output waveform curves for all filter outputs.
2. An automated test method for filters according to claim 1, wherein said ordering of waveform time estimates at different times results in a descending sequence, comprising:
and (5) carrying out descending order arrangement on waveform time evaluations at different time to obtain a descending order sequence.
3. An automated test method for filters according to claim 1, wherein said determining a fluctuation evaluation range from differences in time evaluations of adjacent waveforms in descending order comprises:
calculating the difference value of waveform moment evaluation of two adjacent positions based on the descending sequence, and marking the difference value as an evaluation difference value; recording the waveform moment evaluation with the minimum corresponding sequence in the two waveform moment evaluations corresponding to the maximum evaluation difference as a target evaluation;
and determining a fluctuation evaluation range according to the target evaluation and the maximum waveform moment evaluation.
4. An automated test method for filters according to claim 3, wherein said determining a fluctuation evaluation range from a target evaluation and a maximum waveform moment evaluation comprises:
taking the position corresponding to the maximum waveform moment evaluation in the descending sequence as one end point of the fluctuation evaluation range; and taking the position corresponding to the target evaluation as the other end point of the fluctuation evaluation range, and obtaining the fluctuation evaluation range according to the two end points.
5. The method for automatic testing of a filter according to claim 1, wherein the step of obtaining different to-be-tested waveform segments by cutting out the superimposed waveform curve according to the fluctuation evaluation range includes:
acquiring a plurality of waveform moment evaluations in a fluctuation evaluation range, and recording the waveform moment evaluations as moment evaluations to be selected; evaluating and calculating a middle range value for two adjacent time points to be selected; and intercepting a plurality of different waveform segments to be detected by taking the time sequences corresponding to the two adjacent intermediate range values as endpoints.
6. The method for automatically testing a filter according to claim 5, wherein the analyzing the morphological differences between any two output waveform curve segments corresponding to the waveform segment to be tested to obtain the discrete evaluation between the two output waveform curve segments comprises:
obtaining the minimum matching distance of two output waveform curve segments by a dynamic time warping algorithm on any two output waveform curve segments; obtaining the matching distance of each position in the two output waveform curve sections, and recording the matching distance larger than the minimum matching distance as a larger distance;
the calculation formula of the discrete evaluation is as follows:
the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>For the output waveform curve section->Discrete evaluation of (2); n is the number of the waveform curve segments which are output every two and correspond to the waveform segments to be detected; />For a larger distance->Is the number of (3); />N time points t contained in the current waveform segment X to be detected; />Is a larger distance; />For two output waveform curve sections->Is the minimum matching distance of (2); />The position of the time evaluation to be selected contained in the current waveform segment X to be detected is screened out from the fluctuation evaluation range; />Matching distance for each position of current two output waveform curve sections>Is the maximum position of (2); d is a Euclidean distance function.
7. An automated test method for a filter according to claim 1, wherein determining an overall evaluation of a waveform segment under test from discrete evaluations among all output waveform curve segments corresponding to the waveform segment under test comprises:
and taking the average value of the discrete evaluation among all the output waveform curve sections corresponding to the waveform section to be measured as the integral evaluation of the waveform section to be measured.
8. The automated test method of claim 7, wherein determining the overall evaluation of the waveform segment under test from the discrete evaluations among all of the output waveform curve segments corresponding to the waveform segment under test, further comprises:
if only one output waveform curve section is left, performing discrete evaluation calculation on the output waveform curve section and a corresponding standard waveform curve, and taking the obtained result value as the overall evaluation of the waveform section to be measured.
9. An automated test system for a filter comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of an automated test method for a filter according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311631339.7A CN117330882B (en) | 2023-12-01 | 2023-12-01 | Automatic test method and system for filter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311631339.7A CN117330882B (en) | 2023-12-01 | 2023-12-01 | Automatic test method and system for filter |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117330882A true CN117330882A (en) | 2024-01-02 |
| CN117330882B CN117330882B (en) | 2024-02-06 |
Family
ID=89283399
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311631339.7A Active CN117330882B (en) | 2023-12-01 | 2023-12-01 | Automatic test method and system for filter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117330882B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117890715A (en) * | 2024-03-14 | 2024-04-16 | 大连海事大学 | Filter electrical performance test system based on big data information analysis |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009089490A (en) * | 2007-09-28 | 2009-04-23 | Fujitsu Telecom Networks Ltd | Power supply abnormality detection circuit |
| CN102473593A (en) * | 2009-06-30 | 2012-05-23 | 东京毅力科创株式会社 | Abnormality detection system, abnormality detection method, and recording medium |
| US20140254812A1 (en) * | 2005-09-27 | 2014-09-11 | Ronald Quan | Method and apparatus to evaluate audio equipment via filter banks for dynamic distortions and or differential phase and frequency modulation effects |
| CN105372491A (en) * | 2015-08-31 | 2016-03-02 | 苏州大学 | Method and device for measuring precession frequency |
| CN105866541A (en) * | 2016-06-13 | 2016-08-17 | 公安部第研究所 | Resonance frequency metering method of energy testing simulation card |
| CN107169235A (en) * | 2017-06-14 | 2017-09-15 | 吉林大学 | A kind of multi-parameter collision waveform quality evaluating method |
| US20200149998A1 (en) * | 2018-11-12 | 2020-05-14 | Kabushiki Kaisha Toshiba | Method of detecting anomalies in waveforms, and system thereof |
| US20200278398A1 (en) * | 2017-09-14 | 2020-09-03 | Semiconductor Energy Laboratory Co., Ltd. | Anomaly detection system and anomaly detection method for a secondary battery |
| CN112460492A (en) * | 2020-10-13 | 2021-03-09 | 上海波汇科技有限公司 | Toughness-evaluation-based collaborative toughness-enhanced gas safety control device |
| CN113093272A (en) * | 2021-03-29 | 2021-07-09 | 吉林大学 | Time domain full waveform inversion method based on convolutional coding |
| CN115563880A (en) * | 2022-10-25 | 2023-01-03 | 福建师范大学 | Enterprise power consumption abnormity detection method based on Isolated forest-variable point enhancement |
| WO2023127335A1 (en) * | 2021-12-28 | 2023-07-06 | 株式会社村田製作所 | Abnormality detection device, power supply system, and abnormality detection method |
-
2023
- 2023-12-01 CN CN202311631339.7A patent/CN117330882B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140254812A1 (en) * | 2005-09-27 | 2014-09-11 | Ronald Quan | Method and apparatus to evaluate audio equipment via filter banks for dynamic distortions and or differential phase and frequency modulation effects |
| JP2009089490A (en) * | 2007-09-28 | 2009-04-23 | Fujitsu Telecom Networks Ltd | Power supply abnormality detection circuit |
| CN102473593A (en) * | 2009-06-30 | 2012-05-23 | 东京毅力科创株式会社 | Abnormality detection system, abnormality detection method, and recording medium |
| CN105372491A (en) * | 2015-08-31 | 2016-03-02 | 苏州大学 | Method and device for measuring precession frequency |
| CN105866541A (en) * | 2016-06-13 | 2016-08-17 | 公安部第研究所 | Resonance frequency metering method of energy testing simulation card |
| CN107169235A (en) * | 2017-06-14 | 2017-09-15 | 吉林大学 | A kind of multi-parameter collision waveform quality evaluating method |
| US20200278398A1 (en) * | 2017-09-14 | 2020-09-03 | Semiconductor Energy Laboratory Co., Ltd. | Anomaly detection system and anomaly detection method for a secondary battery |
| US20200149998A1 (en) * | 2018-11-12 | 2020-05-14 | Kabushiki Kaisha Toshiba | Method of detecting anomalies in waveforms, and system thereof |
| CN111175593A (en) * | 2018-11-12 | 2020-05-19 | 株式会社东芝 | Method of detecting anomalies in a waveform and system therefor |
| CN112460492A (en) * | 2020-10-13 | 2021-03-09 | 上海波汇科技有限公司 | Toughness-evaluation-based collaborative toughness-enhanced gas safety control device |
| CN113093272A (en) * | 2021-03-29 | 2021-07-09 | 吉林大学 | Time domain full waveform inversion method based on convolutional coding |
| WO2023127335A1 (en) * | 2021-12-28 | 2023-07-06 | 株式会社村田製作所 | Abnormality detection device, power supply system, and abnormality detection method |
| CN115563880A (en) * | 2022-10-25 | 2023-01-03 | 福建师范大学 | Enterprise power consumption abnormity detection method based on Isolated forest-variable point enhancement |
Non-Patent Citations (3)
| Title |
|---|
| NEILSEN, KD ET AL.: "A stability test for 2-D digital filters", CIRCUIT THEORY AND DESIGN 87. PROCEEDINGS OF THE EUROPEAN CONFERENCE ON CIRCUITS THEORY AND DESIGN - ECCTD 87, pages 807 - 812 * |
| 雷林源: "二维Wiener滤波器及在物探中应用的一些问题", 桂林冶金地质学院学报, no. 1, pages 11 - 20 * |
| 马桂芳 等: "基于 FPGA 和 DSP Builder 的 FIR 数字滤波器设计", 常州工学院学报, vol. 24, no. 5, pages 22 - 26 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117890715A (en) * | 2024-03-14 | 2024-04-16 | 大连海事大学 | Filter electrical performance test system based on big data information analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117330882B (en) | 2024-02-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117330882B (en) | Automatic test method and system for filter | |
| CN109239360B (en) | Reaction curve abnormity detection method and device | |
| CN117889965B (en) | Performance test method of medium-short wave double-color infrared detector | |
| CN108571997A (en) | A kind of method and apparatus that measured point is steadily contacted in detection probe | |
| JP7354421B2 (en) | Error factor estimation device and estimation method | |
| CN116985183A (en) | Quality monitoring and management method and system for near infrared spectrum analyzer | |
| JP6597453B2 (en) | Abnormality diagnosis method, abnormality diagnosis device, and abnormality diagnosis program | |
| KR100259322B1 (en) | Method for analyzing stability of semiconductor of semiconductor device tester | |
| CN116453437B (en) | Display screen module testing method, device, equipment and storage medium | |
| US5111149A (en) | Method and apparatus for automatically calculating the integrity of an electrical coil | |
| CN115376612B (en) | Data evaluation method and device, electronic equipment and storage medium | |
| CN112067971A (en) | VI curve fault phenomenon matrix comparison-based rapid hidden danger detection and diagnosis method | |
| CN116976089A (en) | Robust evaluation method and system for reliability of gas turbine simulation system | |
| CN108536777B (en) | Data processing method, server cluster and data processing device | |
| CN115659271A (en) | Sensor abnormality detection method, model training method, system, device, and medium | |
| JP4235074B2 (en) | Pass / fail judgment device, pass / fail judgment program, and pass / fail judgment method | |
| CN112782233A (en) | Gas identification method based on array gas sensor | |
| JP2008117380A (en) | Method and program for supporting diagnostic process | |
| JP2008078392A (en) | Characteristic analysis method and device, abnormal-equipment estimation method and device, program that allows computer to execute characteristic analysis method or abnormal-equipment estimation method, and computer-readable recording medium that records program | |
| CN110489604A (en) | A kind of analytic method and system for gas turbin test measurement data | |
| US8040958B2 (en) | Method for measuring correlation between frequency response functions | |
| CN116203856B (en) | Universal test method and device based on parameter configuration and storage medium | |
| CN112819751B (en) | Method and device for processing data of detection result of polypeptide chip | |
| KR101518849B1 (en) | Inspection system of connection condition for cable connector and inspection method using the same | |
| CN114580982B (en) | Method, device and equipment for evaluating data quality of industrial equipment |
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 |