A method and system for monitoring mechanical properties of thin film materials

Abstract

The present invention relates to the technical field of mechanical properties of thin film materials, and specifically to a method and system for monitoring the mechanical properties of thin film materials. The present invention identifies image data of a composite thin film divided into a plurality of regions by establishing a thin film defect analysis model; calculates a first monitoring influence value of each defect feature according to the identified defect features, and performs mechanical property monitoring on regions with qualified monitoring of the first monitoring influence value, and divides the regions with unqualified monitoring into a first monitoring interference region; monitors and obtains first performance parameters and second performance parameters of qualified regions, and divides the regions with unqualified monitoring into a second monitoring interference region and a third monitoring interference region; the method can more comprehensively analyze regions in the composite thin film where problems such as uneven stress distribution and defects occur, and provide more accurate composite thin film environmental data, so as to improve the accuracy of mechanical property monitoring by nanoindentation method.

Description

Technical Field

[0001] The invention relates to the technical field of thin film material mechanical properties, and in particular to a thin film material mechanical property monitoring method and system. Background Art

[0002] Film material is a film material with a thickness of less than 1 micron. As a type of film material, plastic film is widely used in packaging and other fields. With the development of technology, people's demand for the performance of plastic film is also increasing, so composite film is born.

[0003] Composite films are synthesized by combining multiple film materials to change the single performance of the original plastic film, obtain the comprehensive performance of multiple film materials, and thus improve the performance of the plastic film; current methods for making composite films include gluing with adhesives or using an extruder to extrude multiple materials from the same die.

[0004] In order to ensure that the composite film can be put into use after production and meet the performance requirements of the usage scenarios, it is necessary to test the performance of the composite film through mechanical property monitoring methods. Direct measurement methods are usually used in the prior art, such as uniaxial tensile tests or uniaxial compression tests, but direct measurement methods are more destructive to film materials. Therefore, the methods currently used are mostly indirect measurement methods such as nanoindentation tests. Compared with direct measurement methods, nanoindentation tests can monitor mechanical properties without destroying the original film material, so as to achieve the purpose of saving costs and multiple tests.

[0005] However, for composite films, there are still some problems in the current nanoindentation method for monitoring the mechanical properties of composite films. Since composite films use a variety of materials, their thickness is thicker than that of traditional film materials. Therefore, uneven stress distribution may occur due to thickness and uniformity issues. At the same time, the microscopic defects and roughness of the composite film material itself will also cause stress concentration. The above results will affect the monitoring results of the mechanical properties, resulting in poor accuracy in mechanical properties monitoring.

[0006] Therefore, a method and system for monitoring the mechanical properties of thin film materials are proposed. Summary of the invention

[0007] The purpose of the present invention is to provide a method and system for monitoring the mechanical properties of thin film materials, by establishing a thin film defect analysis model, identifying image data of a composite film divided into multiple areas; calculating the first monitoring influence value of each defect feature based on the identified defect features, and performing mechanical property monitoring on the area where the first monitoring influence value is qualified, and dividing the area where the monitoring is unqualified into a first monitoring interference area; monitoring and obtaining the first performance parameters and the second performance parameters of the qualified area, and dividing the area where the monitoring is unqualified into a second monitoring interference area and a third monitoring interference area; through this method, it is possible to more comprehensively analyze the areas in the composite film where problems such as uneven stress distribution and defects occur, and provide more accurate composite film environmental data, so as to improve the accuracy of mechanical property monitoring by nanoindentation method.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A method for monitoring the mechanical properties of a thin film material, comprising:

[0010] Acquire image data of a region to be monitored of the composite film, and divide the region to be monitored into a plurality of first monitoring regions; the region to be monitored is divided into regions of size m*m, where m is not less than an indentation influence spacing K;

[0011] Identify and obtain defect features of a plurality of the first monitoring areas according to a thin film defect analysis model, and analyze and calculate a first monitoring influence value of the defect feature in each of the first monitoring areas through the thin film defect analysis model;

[0012] If the first monitoring influence value exceeds the monitoring threshold, the corresponding first monitoring area is divided into a first monitoring interference area; if the first monitoring influence value does not exceed the monitoring threshold, no processing is performed; the mechanical properties of the plurality of the first monitoring areas of the composite film are monitored, and the first performance parameters and the second performance parameters of each of the first monitoring areas are recorded and stored;

[0013] Comparing the first performance parameters of each of the first monitoring areas by using the thin film defect analysis model, and if the comparison result of the first performance parameter is less than a comparison threshold, dividing the corresponding first monitoring area into a second monitoring interference area;

[0014] Further, the film defect analysis model includes a film image input unit, a film characteristic defect feature recognition unit, a film influence value calculation unit, a film performance parameter comparison unit and a film data output unit;

[0015] The film image input unit preprocesses the image data of the first monitoring area to generate preprocessed film data;

[0016] The film feature defect feature recognition unit recognizes the defect feature of the pre-processed film data, marks the defect feature area, and generates film defect feature data; the film feature defect feature recognition unit recognizes the indentation feature of the pre-processed film data, and generates film indentation feature data; the film indentation feature data is the first performance parameter;

[0017] The film influence value calculation unit calculates the first monitoring influence value according to the film defect characteristic data; specifically:

[0018] Traversing all the first monitoring areas, obtaining the thin film defect characteristic data of the i-th first monitoring area, generating a first defect influence parameter according to the area of ​​the largest thin film defect characteristic data of the i-th first monitoring area; generating a second defect influence parameter according to the area of ​​the smallest thin film defect characteristic data of the i-th first monitoring area;

[0019] Generate a third defect influence parameter according to a ratio of the area of ​​all the thin film defect characteristic data in the i-th first monitoring area to the i-th first monitoring area;

[0020] Performing weighted calculation on the first defect impact parameter, the second defect impact parameter, and the third defect impact parameter to obtain the first monitoring impact value;

[0021] The formula is as follows:

[0022] ;

[0023] ; ;

[0024] in, is the first monitoring impact value of the i-th first monitoring area, is the area of ​​the first monitoring area, is the number of defect feature data, is the area of ​​the jth defect feature data of the i-th first monitoring area, is the first defect influencing parameter, is the second defect affecting parameter, is the defect affected area, is the area of ​​all the defect feature data of the i-th first monitoring area, max is the maximum value, min is the minimum value, , and is the defect impact weight, ;

[0025] The film performance parameter comparison unit traverses all the film indentation feature data for comparison, and generates the comparison result of the first performance parameter; specifically:

[0026] Traverse the first performance parameters of all the first monitoring areas, and compare the first performance parameters of the tth first monitoring area with the first performance parameters of the first monitoring area, traversing Second-rate;

[0027] Generating the first performance parameter the comparison results; is the total number of regions of the first performance parameter.

[0028] The formula is as follows:

[0029] ;

[0030] in, is the comparison result of the t-th first performance parameter, is the total number of regions of the first performance parameter, is the kth compared film indentation feature data, is the tth film indentation characteristic data being compared;

[0031] Comparing the second performance parameter of each of the first monitoring areas with reference second performance data, and if the second performance parameter is abnormal, dividing the corresponding first monitoring area into a third monitoring interference area;

[0032] Further, the second performance parameter of each of the first monitoring areas is compared with reference second performance data; specifically:

[0033] Acquire the second performance parameter of each of the first monitoring areas and the trend characteristics of the reference second performance data according to the LSTM model; the second performance parameter includes hardness and elastic modulus;

[0034] Comparing the trend characteristics of the second performance parameter with the reference second performance data, and if an abnormality occurs, obtaining the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter;

[0035] Calculating a comparison result of the first monitoring area according to the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter;

[0036] If the comparison result is less than the abnormal comparison monitoring value, it is determined that the second performance parameter is abnormal;

[0037] The formula is as follows:

[0038] ;

[0039] in, is a comparison result between the second performance parameter of the first monitoring area and the reference second performance data, For the LSTM model, is the second performance parameter, is the reference second performance data, C is the abnormal number statistics, and Calculate weights for anomalies, , the second performance parameter includes hardness and elastic modulus;

[0040] like If the value is less than the abnormal comparison monitoring value, it is judged that the second performance parameter is abnormal;

[0041] The film data output unit outputs the comparison result between the first monitoring impact value and the first performance parameter;

[0042] evaluating the mechanical properties of the composite film according to the first monitoring area, the second monitoring interference area, and the third monitoring interference area;

[0043] The present invention also provides a thin film material mechanical property monitoring system, including a region division module, a first monitoring influence acquisition module, a first monitoring interference region judgment module, a mechanical property monitoring module, a second monitoring interference region judgment module, a third monitoring interference region judgment module and a mechanical property evaluation module, specifically:

[0044] A region division module, which obtains image data of a region to be monitored of the composite film and divides it into a plurality of first monitoring regions;

[0045] A first monitoring impact acquisition module, which identifies and acquires defect features of a plurality of the first monitoring areas according to a thin film defect analysis model, and calculates a first monitoring impact value of the defect feature in each of the first monitoring areas by analyzing the thin film defect analysis model;

[0046] a first monitoring interference area judgment module, which divides the corresponding first monitoring area into a first monitoring interference area if the first monitoring impact value exceeds the monitoring threshold; and does not process if the first monitoring impact value does not exceed the monitoring threshold;

[0047] Furthermore, the calculation and judgment of the first monitoring interference area judgment module is:

[0048] Traversing all the first monitoring areas, obtaining the thin film defect characteristic data of the i-th first monitoring area, generating a first defect influence parameter according to the area of ​​the largest thin film defect characteristic data of the i-th first monitoring area; generating a second defect influence parameter according to the area of ​​the smallest thin film defect characteristic data of the i-th first monitoring area;

[0049] generating a third defect influence parameter according to a ratio of the area of ​​all the thin film defect characteristic data in the i-th first monitoring area to the i-th first monitoring area;

[0050] Performing weighted calculation on the first defect impact parameter, the second defect impact parameter, and the third defect impact parameter to obtain the first monitoring impact value;

[0051] The formula is as follows:

[0052] ;

[0053] ; ;

[0054] in, is the first monitoring impact value of the i-th first monitoring area, is the area of ​​the first monitoring area, is the number of defect feature data, is the area of ​​the jth defect feature data of the i-th first monitoring area, is the first defect influencing parameter, is the second defect affecting parameter, is the defect affected area, is the area of ​​all the defect feature data of the i-th first monitoring area, max is the maximum value, min is the minimum value, , and is the defect impact weight, ;

[0055] A mechanical property monitoring module, which performs mechanical property monitoring on a plurality of the first monitoring regions of the composite film, and records and stores a first performance parameter and a second performance parameter of each of the first monitoring regions;

[0056] a second monitoring interference region judgment module, which compares the first performance parameters of each of the first monitoring regions through the thin film defect analysis model, and if the comparison result of the first performance parameter is less than a comparison threshold, divides the corresponding first monitoring region into a second monitoring interference region;

[0057] Furthermore, the calculation and judgment of the second monitoring interference area judgment module is:

[0058] Traverse the first performance parameters of all the first monitoring areas, and compare the first performance parameters of the tth first monitoring area with the first performance parameters of the first monitoring area, traversing Second-rate;

[0059] Generating the first performance parameter the comparison results; is the total number of regions of the first performance parameter.

[0060] The formula is as follows:

[0061] ;

[0062] in, is the comparison result of the t-th first performance parameter, is the total number of regions of the first performance parameter, is the kth compared film indentation feature data, is the tth film indentation characteristic data being compared;

[0063] a third monitoring interference area judgment module, which compares the second performance parameter of each of the first monitoring areas with the reference second performance data, and if the second performance parameter is abnormal, divides the corresponding first monitoring area into a third monitoring interference area;

[0064] Furthermore, the calculation and judgment of the third monitoring interference area judgment module is:

[0065] Acquire the second performance parameter of each of the first monitoring areas and the trend characteristics of the reference second performance data according to the LSTM model; the second performance parameter includes hardness and elastic modulus;

[0066] Comparing the trend characteristics of the second performance parameter with the reference second performance data, and if an abnormality occurs, obtaining the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter;

[0067] Calculating a comparison result of the first monitoring area according to the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter;

[0068] If the comparison result is less than the abnormal comparison monitoring value, it is determined that the second performance parameter is abnormal;

[0069] The formula is as follows:

[0070] ;

[0071] in, is a comparison result between the second performance parameter of the first monitoring area and the reference second performance data, For the LSTM model, is the second performance parameter, is the reference second performance data, C is the abnormal number statistics, and Calculate weights for anomalies, , the second performance parameter includes hardness and elastic modulus;

[0072] like If the value is less than the abnormal comparison monitoring value, it is judged that the second performance parameter is abnormal;

[0073] A mechanical property evaluation module is used to evaluate the mechanical property of the composite film based on the first monitoring area, the second monitoring interference area and the third monitoring interference area.

[0074] Compared with the prior art, the present invention has the following beneficial effects:

[0075] 1. The present invention identifies image data of a region to be monitored of a composite film based on a film defect analysis model; identifies defect features in each region through the film defect analysis model, and calculates a first monitoring influence value of each defect feature on mechanical monitoring; divides the region into a first monitoring region and a first monitoring interference region, and performs mechanical property monitoring on the first monitoring region; this method can exclude regions with a large number of defect features through quantitative analysis of image data before nanoindentation, so as to avoid the influence of defects and some rough surfaces on the mechanical property monitoring of nanoindentation, thereby improving the accuracy of the monitoring results.

[0076] 2. Based on the nanoindentation results of the first monitoring area, the present invention compares and calculates each nanoindentation feature with all nanoindentation features except itself through a thin film defect analysis model to obtain a comparison result of the nanoindentation features; this method can accurately understand each nanoindentation feature to obtain the depth and size of the nanoindentation, and by comparing with other nanoindentations, it is possible to obtain areas with uneven thickness in the composite film, and by identifying these areas, it is possible to provide accurate composite film environmental data in the nanoindentation mechanical properties analysis, thereby improving the accuracy of the overall composite film mechanical properties analysis.

[0077] 3. In the process of nanoindentation feature comparison, some features may not be obvious enough, resulting in missed uneven areas or defect feature areas in the performance monitoring area. In order to further improve the accuracy of mechanical properties, the LSTM model is used to compare multiple second performance parameters obtained during the mechanical performance monitoring process with the reference second performance data. Based on the comparison, abnormal data is identified to distinguish missed uneven areas or defect feature areas, so as to further reflect more accurate composite film environmental data and improve the accuracy of composite film mechanical property monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0079] Figure 2 It is a schematic diagram of the thin film defect analysis model process of the present invention;

[0080] Figure 3 A schematic diagram of nanoindentation of the composite film of the present invention;

[0081] Figure 4 It is a schematic diagram of the system flow of the present invention. DETAILED DESCRIPTION

[0082] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0083] Since the composite film contains multiple film materials, compared with the performance of the original single film material, the composite film can combine the characteristics of multiple film materials to achieve better performance effects; in the process of monitoring the mechanical properties of the composite film by nanoindentation, due to various reasons such as the production process of the composite film or the properties of the film material, the composite film has some defects or uneven problems, which easily lead to the accuracy of the mechanical property monitoring of nanoindentation. To this end, the present invention provides a method and system for monitoring the mechanical properties of film materials, and the technical solution is as follows:

[0084] Reference Figure 1 As shown, image data of the area to be monitored of the composite film is acquired, and the area to be monitored is divided into a plurality of first monitoring areas;

[0085] Identify and obtain defect features of a plurality of the first monitoring areas according to a thin film defect analysis model, and analyze and calculate a first monitoring influence value of the defect feature in each of the first monitoring areas through the thin film defect analysis model;

[0086] If the first monitoring influence value exceeds the monitoring threshold, the corresponding first monitoring area is divided into a first monitoring interference area; if the first monitoring influence value does not exceed the monitoring threshold, no processing is performed; the mechanical properties of the plurality of the first monitoring areas of the composite film are monitored, and the first performance parameters and the second performance parameters of each of the first monitoring areas are recorded and stored;

[0087] Comparing the first performance parameters of each of the first monitoring areas by using the thin film defect analysis model, and if the comparison result of the first performance parameter is less than a comparison threshold, dividing the corresponding first monitoring area into a second monitoring interference area;

[0088] Comparing the second performance parameter of each of the first monitoring areas with reference second performance data, and if the second performance parameter is abnormal, dividing the corresponding first monitoring area into a third monitoring interference area;

[0089] The mechanical properties of the composite film are evaluated based on the first monitoring area, the second monitoring interference area, and the third monitoring interference area.

[0090] In order to obtain the parameters that affect the accuracy of mechanical performance monitoring in the composite film before the performance test of nanoindentation, the image data of the area to be monitored of the composite film is obtained, and the area to be monitored is divided into multiple areas for regional data quantification, which is convenient for subsequent problem monitoring and mechanical performance testing; improving the accuracy and efficiency of recognition, which is helpful to improve the accuracy of subsequent mechanical performance monitoring;

[0091] The defect characteristics of each area are identified through the film defect analysis model, and all areas are divided into the first monitoring area and the first monitoring interference area through the analysis and calculation of the model. The division avoids the situation in which the mechanical property data obtained by nanoindentation is inaccurate due to defects during the mechanical property monitoring process, so as to improve the accuracy of mechanical property monitoring;

[0092] Based on the first monitoring area, nanoindentation characteristics of each area representing the first performance parameter are obtained, and compared with the nanoindentation characteristics of other areas. Through the comparison results, areas with different characteristics are obtained to identify some areas with different uniformity in the composite film, more accurately understand the non-uniform areas in the composite film, obtain more accurate composite film material environmental data, and improve the accuracy of mechanical property monitoring;

[0093] In order to ensure that these mechanical properties monitoring interference parameters are not missed, after the mechanical properties data monitoring is completed, it is compared with the reference data and analyzed in combination with the time series anomaly model. Through the abnormal analysis results, the missed interference area data is obtained, so as to further obtain accurate composite film material environmental data, further reduce the impact of the interference area data on the mechanical properties monitoring results, and improve the accuracy of mechanical properties monitoring.

[0094] Embodiment 1

[0095] For specific explanation, the present invention is described in conjunction with the following embodiments:

[0096] Obtain image data of the area to be monitored of the composite film, and divide the area to be monitored into multiple first monitoring areas; the area to be monitored is divided into areas of size m*m, where m is not less than the indentation influence spacing K; since the range of the mechanical property monitoring area of ​​the film material is relatively large, in order to be able to quantify the data and improve the accuracy of defect analysis, the image data is divided into multiple areas, and the accuracy of defect identification is improved by narrowing the area range, so as to reduce interference in subsequent calculations and improve the accuracy of mechanical property monitoring; since a trace will appear after indentation, and this trace will have a certain impact on the acquisition of the next mechanical property data, in order not to affect the mechanical property influence of the distance between indentations during the nanoindentation process, the area is set not to be less than the indentation influence spacing K, and the value of K is obtained based on the current indentation test environment, combined with a large amount of historical data and expert experience;

[0097] Identify and obtain defect features of a plurality of the first monitoring areas according to a thin film defect analysis model, and analyze and calculate a first monitoring influence value of the defect feature in each of the first monitoring areas through the thin film defect analysis model;

[0098] If the first monitoring influence value exceeds the monitoring threshold, the corresponding first monitoring area is divided into a first monitoring interference area; if the first monitoring influence value does not exceed the monitoring threshold, no processing is performed; the mechanical properties of the plurality of the first monitoring areas of the composite film are monitored, and the first performance parameters and the second performance parameters of each of the first monitoring areas are recorded and stored; the monitoring threshold is set based on the first monitoring influence value calculated from a large amount of historical experimental data, and from the first monitoring influence values ​​obtained from these historical data, in combination with expert experience and actual experimental scenarios;

[0099] The first performance parameters of each first monitoring area are compared by the thin film defect analysis model. If the comparison result of the first performance parameter is less than the comparison threshold, the corresponding first monitoring area is divided into a second monitoring interference area. The comparison threshold and the monitoring threshold are similar. The comparison result of the first performance parameter calculated based on the first performance parameter of a large amount of historical experimental data is 0.95 by default, and can be set in combination with expert experience and actual experimental scenarios.

[0100] Furthermore, the film defect analysis model includes a film image input unit, a film characteristic defect feature recognition unit, a film influence value calculation unit, a film performance parameter comparison unit and a film data output unit; Figure 2 As shown;

[0101] The film image input unit preprocesses the image data of the first monitoring area to generate preprocessed film data;

[0102] The film feature defect feature recognition unit recognizes the defect features of the pre-processed film data, marks the defect feature area, and generates film defect feature data; the film feature defect feature recognition unit recognizes the indentation features of the pre-processed film data, and generates film indentation feature data; the film indentation feature data is the first performance parameter; the film indentation feature data adopts the ResNet model, which can quickly learn and extract information features in the image data, and at the same time, the identified defect feature results can provide a reliable data basis for the subsequent calculation of the first monitoring influence value, which helps to reduce the interference of defects of the composite film on the mechanical property monitoring of nanoindentation and improve the accuracy of mechanical property monitoring;

[0103] This embodiment obtains some experimental data, and compares the defect features identified by some current deep image recognition models. The experimental results of the recognition accuracy are shown in Table 1:

[0104] Table 1 Comparison of film defect feature recognition results of multiple models

[0105]

[0106] The film influence value calculation unit calculates the first monitoring influence value according to the film defect characteristic data; specifically:

[0107] Traversing all the first monitoring areas, obtaining the thin film defect characteristic data of the i-th first monitoring area, generating a first defect influence parameter according to the area of ​​the largest thin film defect characteristic data of the i-th first monitoring area; generating a second defect influence parameter according to the area of ​​the smallest thin film defect characteristic data of the i-th first monitoring area;

[0108] Generate a third defect influence parameter according to a ratio of the area of ​​all the thin film defect characteristic data in the i-th first monitoring area to the i-th first monitoring area;

[0109] Performing weighted calculation on the first defect impact parameter, the second defect impact parameter, and the third defect impact parameter to obtain the first monitoring impact value;

[0110] The formula is as follows:

[0111] ;

[0112] ; ;

[0113] in, is the first monitoring impact value of the i-th first monitoring area, is the area of ​​the first monitoring area, is the number of defect feature data, is the area of ​​the jth defect feature data of the i-th first monitoring area, is the first defect influencing parameter, is the second defect affecting parameter, is the defect affected area, is the area of ​​all the defect feature data of the i-th first monitoring area, max is the maximum value, min is the minimum value, , and is the defect impact weight, ; The default value is 0.5. and The default values ​​are 0.25 respectively; they can be modified according to the actual situation;

[0114] In the actual generation process, although the defects on the composite film have a certain influence on the nanoindentation, the different distances and sizes will have different degrees of influence on the test in the divided area. In order to determine the impact of the identified defect characteristics on the data of mechanical property monitoring, the identified defect feature area is calculated, and the identified defect feature area data is quantitatively mapped into evaluation data. Judging by the evaluation data can help relevant personnel determine whether the area can be used for mechanical property monitoring, thereby improving the accuracy of the composite film nanoindentation test;

[0115] The film performance parameter comparison unit traverses all the film indentation feature data for comparison, and generates the comparison result of the first performance parameter; referring to Figure 3 As shown, the triangle in the box is the film indentation after identification; specifically:

[0116] Traverse the first performance parameters of all the first monitoring areas, and compare the first performance parameters of the tth first monitoring area with the first performance parameters of the first monitoring area, traversing Second-rate;

[0117] Generating the first performance parameter the comparison results; is the total number of regions of the first performance parameter.

[0118] The formula is as follows:

[0119] ;

[0120] in, is the comparison result of the t-th first performance parameter, is the total number of regions of the first performance parameter, is the kth compared film indentation feature data, is the tth film indentation characteristic data being compared;

[0121] In the mechanical property monitoring of the nanoindentation test, since the uniformity of the film cannot be significantly analyzed through image data, it cannot be accurately identified. Therefore, after the nanoindentation test, all the nanoindentation features are traversed and compared with the nanoindentation features of other areas. Through the similarity of the image data, it can be determined whether the area has a problem of uneven thickness. This method can quantify the unevenness problem, and indirectly reflect the uneven thickness problem in the film through the data in the indentation test. The specific situation of the film material can be more accurately understood to improve the accuracy of nanoindentation mechanical property monitoring;

[0122] Compare the second performance parameter of each first monitoring area with the reference second performance data. If the second performance parameter is abnormal, divide the corresponding first monitoring area into a third monitoring interference area. The reference second performance data is set based on a large amount of historical data. It includes LSTM model and comparison calculation.

[0123] Further, the film data output unit includes an LSTM model and a comparison calculation unit, the LSTM model is used to identify anomalies between the second performance parameter and the reference second performance data, and the comparison calculation unit compares the second performance parameter of each first monitoring area with the reference second performance data as follows:

[0124] Acquire the second performance parameter of each of the first monitoring areas and the trend characteristics of the reference second performance data according to the LSTM model; the second performance parameter includes hardness and elastic modulus;

[0125] Comparing the trend characteristics of the second performance parameter with the reference second performance data, and if an abnormality occurs, obtaining the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter;

[0126] Calculating a comparison result of the first monitoring area according to the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter;

[0127] If the comparison result is less than the abnormal comparison monitoring value, it is determined that the second performance parameter is abnormal;

[0128] The formula is as follows:

[0129] ;

[0130] in, is a comparison result between the second performance parameter of the first monitoring area and the reference second performance data, For the LSTM model, is the second performance parameter, is the reference second performance data, C is the abnormal number statistics, and Calculate weights for anomalies, , the second performance parameter includes hardness and elastic modulus;

[0131] The LSTM model uses a 4-layer LSTM unit, is trained by a large amount of historical data annotated with abnormal labels, obtains corresponding model parameters, and generates the LSTM model;

[0132] During the identification process, some defects and uneven problems may be missed. Therefore, the hardness data and elastic modulus data obtained through multiple nanoindentation tests are set at fixed time intervals to form time series data. The LSTM model is used to compare and mine the abnormal characteristics of the hardness data and elastic modulus data hidden in the time series data. These data can be used to find out the missed defects and uneven problems, and can be provided to relevant technical personnel for reference to understand the actual situation of the composite film, reduce interference with the monitoring results of the mechanical properties of the composite film, and improve the accuracy of the monitoring of the mechanical properties of the composite film.

[0133] like If the value is less than the abnormal comparison monitoring value, the second performance parameter is judged to be abnormal; the comparison monitoring value is set based on a large amount of historical experimental data combined with the result of formula calculation, through expert experience and actual experimental scenarios;

[0134] The film data output unit outputs the comparison result between the first monitoring impact value and the first performance parameter;

[0135] evaluating the mechanical properties of the composite film according to the first monitoring area, the second monitoring interference area, and the third monitoring interference area;

[0136] Embodiment 2

[0137] The present invention also provides a thin film material mechanical property monitoring system, which is based on a thin film material mechanical property monitoring method, inputs the acquired data into a corresponding model for processing, and can quickly feedback to relevant technical personnel, avoiding the influence of defects and unevenness on the mechanical property monitoring of the composite film in the nanoindentation test, and improving the accuracy of the mechanical property monitoring results. The system includes a region division module, a first monitoring influence acquisition module, a first monitoring interference region judgment module, a mechanical property monitoring module, a second monitoring interference region judgment module, a third monitoring interference region judgment module and a mechanical property evaluation module, with reference to Figure 4 As shown, specifically:

[0138] A region division module, which obtains image data of a region to be monitored of the composite film and divides it into a plurality of first monitoring regions;

[0139] A first monitoring impact acquisition module, which identifies and acquires defect features of a plurality of the first monitoring areas according to a thin film defect analysis model, and calculates a first monitoring impact value of the defect feature in each of the first monitoring areas by analyzing the thin film defect analysis model;

[0140] a first monitoring interference area judgment module, which divides the corresponding first monitoring area into a first monitoring interference area if the first monitoring impact value exceeds the monitoring threshold; and does not process if the first monitoring impact value does not exceed the monitoring threshold;

[0141] Furthermore, the calculation and judgment of the first monitoring interference area judgment module is:

[0142] Traversing all the first monitoring areas, obtaining the thin film defect characteristic data of the i-th first monitoring area, generating a first defect influence parameter according to the area of ​​the largest thin film defect characteristic data of the i-th first monitoring area; generating a second defect influence parameter according to the area of ​​the smallest thin film defect characteristic data of the i-th first monitoring area;

[0143] Generate a third defect influence parameter according to a ratio of the area of ​​all the thin film defect characteristic data in the i-th first monitoring area to the i-th first monitoring area;

[0144] Performing weighted calculation on the first defect impact parameter, the second defect impact parameter, and the third defect impact parameter to obtain the first monitoring impact value;

[0145] The formula is as follows:

[0146] ;

[0147] ; ;

[0148] in, is the first monitoring impact value of the i-th first monitoring area, is the area of ​​the first monitoring area, is the number of defect feature data, is the area of ​​the jth defect feature data of the i-th first monitoring area, is the first defect influencing parameter, is the second defect affecting parameter, is the defect affected area, is the area of ​​all the defect feature data of the i-th first monitoring area, max is the maximum value, min is the minimum value, , and is the defect impact weight, ;

[0149] A mechanical property monitoring module, which performs mechanical property monitoring on a plurality of the first monitoring regions of the composite film, and records and stores a first performance parameter and a second performance parameter of each of the first monitoring regions;

[0150] a second monitoring interference region judgment module, which compares the first performance parameters of each of the first monitoring regions through the thin film defect analysis model, and if the comparison result of the first performance parameter is less than a comparison threshold, divides the corresponding first monitoring region into a second monitoring interference region;

[0151] Furthermore, the calculation and judgment of the second monitoring interference area judgment module is:

[0152] Traverse the first performance parameters of all the first monitoring areas, and compare the first performance parameters of the tth first monitoring area with the first performance parameters of the first monitoring area, traversing Second-rate;

[0153] Generating the first performance parameter the comparison results; is the total number of regions of the first performance parameter.

[0154] The formula is as follows:

[0155] ;

[0156] in, is the comparison result of the t-th first performance parameter, is the total number of regions of the first performance parameter, is the kth compared film indentation feature data, is the tth film indentation characteristic data being compared;

[0157] a third monitoring interference area judgment module, which compares the second performance parameter of each of the first monitoring areas with the reference second performance data, and if the second performance parameter is abnormal, divides the corresponding first monitoring area into a third monitoring interference area;

[0158] Furthermore, the calculation and judgment of the third monitoring interference area judgment module is:

[0159] Acquire the second performance parameter of each of the first monitoring areas and the trend characteristics of the reference second performance data according to the LSTM model; the second performance parameter includes hardness and elastic modulus;

[0160] Comparing the trend characteristics of the second performance parameter with the reference second performance data, and if an abnormality occurs, obtaining the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter;

[0161] Calculating a comparison result of the first monitoring area according to the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter;

[0162] If the comparison result is less than the abnormal comparison monitoring value, it is determined that the second performance parameter is abnormal;

[0163] The formula is as follows:

[0164] ;

[0165] in, is a comparison result between the second performance parameter of the first monitoring area and the reference second performance data, For the LSTM model, is the second performance parameter, is the reference second performance data, C is the abnormal number statistics, and Calculate weights for anomalies, , the second performance parameter includes hardness and elastic modulus;

[0166] like If the value is less than the abnormal comparison monitoring value, it is judged that the second performance parameter is abnormal;

[0167] A mechanical property evaluation module is used to evaluate the mechanical property of the composite film based on the first monitoring area, the second monitoring interference area and the third monitoring interference area.

[0168] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring the mechanical properties of thin film materials, characterized in that: include: Acquire image data of a region to be monitored of the composite film, and divide the region to be monitored into a plurality of first monitoring regions; Identify and obtain defect features of a plurality of the first monitoring areas according to a thin film defect analysis model, and analyze and calculate a first monitoring influence value of the defect feature in each of the first monitoring areas through the thin film defect analysis model; If the first monitoring impact value exceeds the monitoring threshold, the corresponding first monitoring area is divided into a first monitoring interference area; if the first monitoring impact value does not exceed the monitoring threshold, no processing is performed; Performing mechanical property monitoring on a plurality of the first monitoring regions of the composite film, and recording and storing a first performance parameter and a second performance parameter of each of the first monitoring regions; Comparing the first performance parameters of each of the first monitoring areas by using the thin film defect analysis model, and if the comparison result of the first performance parameter is less than a comparison threshold, dividing the corresponding first monitoring area into a second monitoring interference area; Comparing the second performance parameter of each of the first monitoring areas with reference second performance data, and if the second performance parameter is abnormal, dividing the corresponding first monitoring area into a third monitoring interference area; The mechanical properties of the composite film are evaluated based on the first monitoring area, the second monitoring interference area, and the third monitoring interference area.

2. A method for monitoring the mechanical properties of thin film materials according to claim 1, characterized in that: The area to be monitored is divided into areas of size m*m, where m is not less than the indentation influence spacing K.

3. A method for monitoring the mechanical properties of thin film materials according to claim 1, characterized in that: The thin film defect analysis model identifies defect features of a plurality of first monitoring areas, calculates first monitoring influence values ​​of the defect features in each first monitoring area, and compares the first performance parameters of each first monitoring area, including: The film defect analysis model includes a film image input unit, a film characteristic defect feature recognition unit, a film influence value calculation unit, a film performance parameter comparison unit and a film data output unit; The film image input unit preprocesses the image data of the first monitoring area to generate preprocessed film data; The film feature defect feature recognition unit recognizes the defect feature of the pre-processed film data, marks the defect feature area, and generates film defect feature data; the film feature defect feature recognition unit recognizes the indentation feature of the pre-processed film data, and generates film indentation feature data; the film indentation feature data is the first performance parameter; The film influence value calculation unit calculates the first monitoring influence value based on the film defect characteristic data; the film performance parameter comparison unit traverses all the film indentation characteristic data for comparison, and generates the comparison result of the first performance parameter; the film data output unit outputs the comparison result of the first monitoring influence value and the first performance parameter.

4. A method for monitoring the mechanical properties of thin film materials according to claim 3, characterized in that: The thin film influence value calculation unit calculates the first monitoring influence value according to the thin film defect characteristic data, including: Traversing all the first monitoring areas, obtaining the thin film defect characteristic data of the i-th first monitoring area, generating a first defect influence parameter according to the area of ​​the largest thin film defect characteristic data of the i-th first monitoring area; generating a second defect influence parameter according to the area of ​​the smallest thin film defect characteristic data of the i-th first monitoring area; generating a third defect influence parameter according to a ratio of the area of ​​all the thin film defect characteristic data in the i-th first monitoring area to the i-th first monitoring area; The first defect impact parameter, the second defect impact parameter and the third defect impact parameter are weightedly calculated to obtain the first monitoring impact value.

5. A method for monitoring the mechanical properties of thin film materials according to claim 3, characterized in that: The film performance parameter comparison unit traverses all the film indentation feature data for comparison, and generates the comparison result of the first performance parameter, including: Traverse the first performance parameters of all the first monitoring areas, and compare the first performance parameters of the tth first monitoring area with the first performance parameters of the first monitoring area, traversing Second-rate; Generating the first performance parameter the comparison results; is the total number of regions of the first performance parameter.

6. A method for monitoring the mechanical properties of thin film materials according to claim 1, characterized in that: Comparing the second performance parameter of each of the first monitoring areas with reference second performance data includes: Acquire the second performance parameter of each of the first monitoring areas and the trend characteristics of the reference second performance data according to the LSTM model; the second performance parameter includes hardness and elastic modulus; Comparing the trend characteristics of the second performance parameter with those of the reference second performance data, and if an abnormality occurs, obtaining the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter; Calculating a comparison result of the first monitoring area according to the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter; If the comparison result is less than the abnormal comparison monitoring value, it is determined that the second performance parameter is abnormal.

7. A film material mechanical properties monitoring system, characterized in that: include: A region division module, which obtains image data of a region to be monitored of the composite film and divides it into a plurality of first monitoring regions; A first monitoring impact acquisition module, which identifies and acquires defect features of a plurality of the first monitoring areas according to a thin film defect analysis model, and calculates a first monitoring impact value of the defect feature in each of the first monitoring areas; a first monitoring interference area judgment module, which divides the corresponding first monitoring area into a first monitoring interference area if the first monitoring impact value exceeds the monitoring threshold; and does not process if the first monitoring impact value does not exceed the monitoring threshold; A mechanical property monitoring module, which performs mechanical property monitoring on a plurality of the first monitoring regions of the composite film, and records and stores a first performance parameter and a second performance parameter of each of the first monitoring regions; a second monitoring interference region judgment module, which compares the first performance parameters of each of the first monitoring regions through the thin film defect analysis model, and if the comparison result of the first performance parameter is less than a comparison threshold, divides the corresponding first monitoring region into a second monitoring interference region; a third monitoring interference area judgment module, which compares the second performance parameter of each of the first monitoring areas with the reference second performance data, and if the second performance parameter is abnormal, divides the corresponding first monitoring area into a third monitoring interference area; A mechanical property evaluation module is used to evaluate the mechanical property of the composite film based on the first monitoring area, the second monitoring interference area and the third monitoring interference area.

8. A thin film material mechanical property monitoring system according to claim 7, characterized in that: The calculation and judgment of the first monitoring interference area judgment module includes: Traversing all the first monitoring areas, obtaining thin film defect characteristic data of the i-th first monitoring area, generating a first defect influence parameter according to the area of ​​the largest thin film defect characteristic data of the i-th first monitoring area; generating a second defect influence parameter according to the area of ​​the smallest thin film defect characteristic data of the i-th first monitoring area; generating a third defect influence parameter according to a ratio of the area of ​​all the thin film defect characteristic data in the i-th first monitoring area to the i-th first monitoring area; The first defect impact parameter, the second defect impact parameter and the third defect impact parameter are weightedly calculated to obtain the first monitoring impact value.

9. A film material mechanical property monitoring system according to claim 7, characterized in that: The calculation and judgment of the second monitoring interference area judgment module includes: Traverse the first performance parameters of all the first monitoring areas, and compare the first performance parameters of the tth first monitoring area with the first performance parameters of the first monitoring area, traversing Second-rate; Generating the first performance parameter the comparison results; is the total number of regions of the first performance parameter.

10. A thin film material mechanical property monitoring system according to claim 7, characterized in that: The calculation and judgment of the third monitoring interference area judgment module includes: Acquire the second performance parameter of each of the first monitoring areas and the trend characteristics of the reference second performance data according to the LSTM model; the second performance parameter includes hardness and elastic modulus; Comparing the trend characteristics of the second performance parameter with those of the reference second performance data, and if an abnormality occurs, obtaining the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter; The comparison result of the first monitoring area is calculated according to the abnormal trend characteristic trend degree and the abnormal trend characteristic trend quantity of the second performance parameter; if the comparison result is less than the abnormal comparison monitoring value, the second performance parameter is judged to be abnormal.

Images