Method and system for optimizing interfacial strength of ink-jet printing film sensor

Abstract

According to the method and the system for optimizing the interface strength of the ink-jet printing film sensor, provided by the invention, the meso-structure characteristics of the ink-jet printing film sensor with different process parameters under a multi-dimensional process type are observed and acquired, and the LSTM neural network is trained by an ink-jet process parameter matrix and the three-dimensional meso-structure characteristics; obtaining a trained neural network model; a load displacement curve of a scribing head is calculated and obtained through a preset micro-mechanical model and a phase field fracture model, the real interface bonding strength is obtained, the real interface bonding strength is introduced, the trained neural network model is trained again, and a process parameter-micro-structure-interface bonding strength coupling structure-activity relationship is obtained; and in combination with a genetic algorithm, optimizing the bonding strength of the current interface according to the corresponding process parameters in the ink jet process parameter matrix. The parameter optimization period is shortened, the interface strength optimization cost is reduced, and the optimization efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of metal processing, and in particular to a method and system for optimizing the interface strength of an inkjet-printed thin film sensor. Background Art

[0002] Inkjet printing technology has been widely used in sensor manufacturing in recent years, especially thin film sensors. However, the performance of thin film sensors is significantly affected by their interface strength. The interface strength is not only related to the reliability and durability of the sensor, but also affects its sensitivity and response time. Therefore, the research on the interface strength control technology of inkjet printed thin films has gradually received attention.

[0003] At present, the interface strength control technology of inkjet printed thin film sensors mainly focuses on the following aspects: material selection and modification; surface treatment technology; printing parameter optimization; post-processing process; and multi-layer structure design.

[0004] In the existing inkjet printing thin film sensor interface strength control technology, the main exploration is based on the "trial and error" experimental matrix formed by orthogonal preparation process parameters. The disadvantage of this technical method is that when the process types and parameters increase, the scale of the experimental matrix will explode exponentially, resulting in a sharp increase in experimental costs and a longer process optimization cycle.

[0005] Taking the process that has a greater impact on the interface strength of inkjet printed thin film sensors as an example, there are 6 groups of process types, including printing speed, printing spacing, substrate temperature, laser sintering power, laser scanning spacing, and laser spot diameter. Each group of process types contains 10 to 20 groups of process parameters, and the elements in its orthogonal test matrix can reach 10 6 Up to 2×10 6 On this basis, carrying out detailed thin film interface strength test requires a lot of manpower and material resources, and every time the thin film sensor material system changes, the above huge amount of test work needs to be repeated, which is extremely difficult to implement.

[0006] Therefore, it is an urgent problem to be solved by those skilled in the art to provide a method and system for optimizing the interface strength of an inkjet-printed thin film sensor for solving the above-mentioned problems. Summary of the invention

[0007] The purpose of the present invention is to provide an inkjet-printed thin film sensor interface strength optimization method, which has clear logic, is safe, effective, reliable and easy to operate, and can effectively reduce the cost of inkjet-printed thin film sensor interface strength optimization and improve optimization efficiency.

[0008] Based on the above objectives, the technical solution provided by the present invention is as follows: A method for optimizing the interface strength of an inkjet printed thin film sensor comprises the following steps: Provide inkjet printed thin film sensors with different process parameters under multi-dimensional process types; Observe and obtain the microstructure map of the inkjet printed thin film sensor, and extract the three-dimensional microstructure characteristics of the inkjet printed thin film sensor after processing; An inkjet process parameter matrix formed by combining different process parameters under multi-dimensional process types is used as an input variable, and a three-dimensional microstructure feature corresponding to the inkjet process parameter matrix is ​​used as a target to train the LSTM neural network to obtain a trained neural network model; According to the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix, the preset inkjet printing thin film sensor micromechanics model and the phase field fracture model, the load displacement curve of the scratching head is calculated and obtained, and the real interface bonding strength is obtained according to the load displacement curve of the scratching head; Taking the inkjet process parameter matrix as input variables, taking the three-dimensional microstructure characteristics and the interface bonding strength corresponding to the inkjet process parameter matrix as outputs, the trained neural network model is trained again to obtain the process parameter-microstructure-interface bonding strength coupled structure-activity relationship; According to the process parameter-microstructure-interface bonding strength coupled structure-activity relationship and the genetic algorithm, the corresponding process parameters are searched in the inkjet process parameter matrix to optimize the current interface bonding strength.

[0009] Preferably, the observing and obtaining of the microstructure map of the inkjet printed thin film sensor and extracting the three-dimensional microstructure features of the inkjet printed thin film sensor after processing comprises the following steps: Observe the inkjet-printed thin film sensors with different process parameters under various multi-dimensional process types to obtain corresponding various micro-structure maps; According to an adaptive threshold algorithm, the mesoscopic tissue structure map is converted into a binary tissue map; Based on stereological methods, the tissue atlas is converted into three-dimensional microscopic tissue structural features.

[0010] Preferably, a preset mesoscopic mechanical model and phase field fracture model of an inkjet-printed thin film sensor are obtained, specifically: the mesoscopic mechanical model of the inkjet-printed thin film sensor is constructed in abaqus finite element software according to Python script programming; The phase field fracture model is constructed using the uel program in the abaqus finite element software.

[0011] Preferably, the load displacement curve of the scratching head is calculated based on the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix, the preset inkjet printing thin film sensor micromechanics model and the phase field fracture model, and the real interface bonding strength is obtained according to the load displacement curve of the scratching head, including the following steps: Inputting the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix into the micromechanical model of the inkjet printed thin film sensor to obtain the stress value corresponding to the inkjet process parameter matrix; Inputting the stress value corresponding to the inkjet process parameter matrix into the phase field fracture model, obtaining the moving distance of the indenter and the vertical load change, and integrating to obtain the load displacement curve of the scratching head; It is determined whether the load displacement curve of the scratching head meets a preset condition. If so, the inkjet process parameter matrix is ​​traversed to calculate the interface bonding strength as the real interface bonding strength.

[0012] Preferably, the process parameter-microstructure-interface bonding strength coupling structure-activity relationship is specifically: There is a mapping relationship between the inkjet process parameter matrix and the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix; The three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix and the interface bonding strength are in a power function relationship, and the specific formula is: ; in, is the interface bonding strength, It is the three-dimensional microstructure feature. and is the weight coefficient.

[0013] Preferably, according to the process parameter-microstructure-interface bonding strength coupling structure-activity relationship and genetic algorithm, the corresponding process parameters in the inkjet process parameter matrix are used to optimize the current interface bonding strength, specifically: Determining optimal process parameters in the inkjet process parameter matrix according to the genetic algorithm; The optimized interface bonding strength is determined according to the optimal process parameters, and the current interface strength of the inkjet printed thin film sensor is updated with the optimized interface bonding strength.

[0014] An inkjet printed thin film sensor interface strength optimization system, comprising: A three-dimensional microstructure feature module is used to observe and obtain the microstructure map of the inkjet-printed thin film sensor, and extract the three-dimensional microstructure features of the inkjet-printed thin film sensor after processing; A neural network training module is used to form an inkjet process parameter matrix formed by combining different process parameters under multi-dimensional process types as input variables, and to train the LSTM neural network with the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix as a target to obtain a trained neural network model; A real interface bonding strength module is used to calculate and obtain the load displacement curve of the scratching head according to the three-dimensional microscopic structural characteristics corresponding to the inkjet process parameter matrix, the preset inkjet printing thin film sensor microscopic mechanical model and the phase field fracture model, and obtain the real interface bonding strength according to the load displacement curve of the scratching head; A structure-activity relationship module is used to use the inkjet process parameter matrix as an input variable, and the three-dimensional microstructure characteristics and the interface binding strength corresponding to the inkjet process parameter matrix as outputs, and to train the trained neural network model again to obtain the process parameter-microstructure-interface binding strength coupled structure-activity relationship; The interface bonding strength optimization module optimizes the current interface bonding strength according to the process parameter-microstructure-interface bonding strength coupling structure-activity relationship and the genetic algorithm, and the corresponding process parameters in the inkjet process parameter matrix.

[0015] The interface strength optimization method of the inkjet-printed thin film sensor provided by the present invention is to observe and obtain the microstructure characteristics of the inkjet-printed thin film sensor with different process parameters under multi-dimensional process types, train the LSTM neural network with the inkjet process parameter matrix and the three-dimensional microstructure characteristics to obtain the trained neural network model; by presetting the micromechanical model and the phase field fracture model, calculate and obtain the load displacement curve of the scratching head to obtain the real interface bonding strength, introduce the real interface bonding strength, and train the trained neural network model again to obtain the process parameter-microstructure-interface bonding strength coupling structure-activity relationship; combined with the genetic algorithm, the corresponding process parameters in the inkjet process parameter matrix are used to optimize the current interface bonding strength.

[0016] Compared with the prior art, the present invention introduces the LSTM machine learning method, obtains the coupled structure-activity relationship through digital simulation, and performs targeted optimization of the inkjet printing process parameters in combination with the interface strength, thus overcoming the need for a large number of mechanical experiments in the original optimization process, shortening the parameter optimization cycle, reducing the cost of interface strength optimization, and improving the optimization efficiency.

[0017] The present invention also provides an inkjet-printed thin film sensor interface strength optimization system. Since it has the same technical concept as this method, solves the same technical problem, and should have the same beneficial effects, it will not be described in detail here. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flow chart of a method for optimizing the interface strength of an inkjet-printed thin film sensor provided by an embodiment of the present invention; Figure 2 A flowchart of step S2 provided in an embodiment of the present invention; Figure 3 A flowchart of step S4 provided in an embodiment of the present invention; Figure 4 A flowchart of step S6 provided in an embodiment of the present invention; Figure 5 A schematic structural diagram of an inkjet-printed thin film sensor interface strength optimization system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] 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.

[0021] The embodiments of the present invention are written in a progressive manner.

[0022] The embodiment of the present invention provides an inkjet printed thin film sensor interface strength optimization method and system, which mainly solves the technical problem that in the prior art, detailed thin film interface strength test experiments require a lot of manpower and material resources, and whenever the thin film sensor material system changes, the above huge amount of test work needs to be repeated, which is extremely difficult to perform.

[0023] like Figure 1 As shown, a method for optimizing the interface strength of an inkjet printed thin film sensor comprises the following steps: S1. Provide inkjet printed thin film sensors with different process parameters under multi-dimensional process types; S2. Observe and obtain the microstructure map of the inkjet-printed thin film sensor, and extract the three-dimensional microstructure characteristics of the inkjet-printed thin film sensor after processing; S3. Taking the inkjet process parameter matrix formed by combining different process parameters under multi-dimensional process types as input variables, taking the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix as the target, training the LSTM neural network to obtain a trained neural network model; S4. According to the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix, the preset inkjet printed thin film sensor micromechanics model and the phase field fracture model, the load displacement curve of the scratching head is calculated and obtained, and the real interface bonding strength is obtained according to the load displacement curve of the scratching head; S5. Taking the inkjet process parameter matrix as input variables and the three-dimensional microstructure characteristics and interface bonding strength corresponding to the inkjet process parameter matrix as outputs, the trained neural network model is trained again to obtain the process parameter-microstructure-interface bonding strength coupled structure-activity relationship; S6. According to the coupled structure-activity relationship of process parameters-microstructure-interface bonding strength and genetic algorithm, the corresponding process parameters are searched in the inkjet process parameter matrix to optimize the current interface bonding strength.

[0024] In step S1, an inkjet printed thin film sensor with different process parameters under process types including printing speed, printing spacing, substrate temperature, laser sintering power, laser scanning spacing, and laser spot diameter is provided. 5 to 10 process parameters are selected under each process type to ensure the versatility of the process optimization path and the effectiveness of the test results, as shown in Table 1: Table 1. Different process parameters of inkjet printed thin film sensors under multi-dimensional process types

[0025] In step S2, the structure of the inkjet printed thin film sensor is observed to obtain a microscopic tissue map, and after processing, the three-dimensional microscopic tissue structure characteristics of various inkjet printed thin film sensors are proposed; In step S3, the inkjet process parameter matrix formed by combining different process parameters under multi-dimensional process types is used as the input variable X, and the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix are used as the target Y to train the LSTM neural network to obtain a trained neural network model; It should be noted that the Long Short-Term Memory (LSTM) network is a time recurrent neural network designed to solve the long-term dependency problem of general RNNs (recurrent neural networks). All RNNs have a chain form of repeated neural network modules. In standard RNNs, this repeated structural module has only a very simple structure, such as a tanh layer.

[0026] In step S4, the load displacement curve of the scratching head is calculated according to the three-dimensional microstructure characteristics in combination with the preset micromechanical model of the printed thin film sensor and the phase field fracture model, thereby obtaining the real interface bonding strength; In this embodiment, the critical load is determined by a micron scratch test. The scratch width is determined by observing the scratch morphology using an optical microscope. The scratch depth is determined by the drop distance of the displacement sensor at the top of the indenter. The elastic modulus of the film is determined by a nanoindentation test. The true interface bonding strength can be calculated using these four parameters.

[0027] In step S5, the inkjet process parameter matrix is ​​used as the input variable X, and the three-dimensional microstructure characteristics and interface bonding strength corresponding to the inkjet process parameter matrix are used as the output Y, and the neural network model trained in step S3 is trained again to obtain the process parameter-microstructure-interface bonding strength coupled structure-activity relationship; In this embodiment, taking the process parameter porosity as an example, the structure-activity relationship of porosity is due to the different porosities on the surface of the ITO film under different process parameters. By analyzing and counting the microstructure maps of different process parameters, the porosity under different process parameters can be obtained.

[0028] In step S6, a genetic algorithm is introduced to combine the process parameter-microstructure-interface bonding strength coupling structure-activity relationship, and the corresponding process parameters are searched in the inkjet process parameter matrix to optimize the current interface bonding strength.

[0029] In this embodiment, there are five process parameters for each of the six process types. First, determine a process type, such as laser power, and select the power with the largest interface bonding strength among the five laser powers. Determine this power as the optimal power, and then modify another process type, such as printing speed. After determining the optimal power, compare the five scanning speeds and select the scanning speed corresponding to the largest bonding strength as the optimal scanning speed. After determining the optimal power and the optimal speed, select a new process type, and so on, and finally determine the optimal interface strength under the six process parameters.

[0030] like Figure 2 As shown, preferably, step S2 includes the following steps: A1. Observe the inkjet-printed thin film sensors with different process parameters under various multi-dimensional process types to obtain the corresponding various micro-structure maps; A2. Convert the mesoscopic tissue structure map into a binary tissue map according to an adaptive threshold algorithm; A3. Based on stereological methods, the tissue atlas is converted into three-dimensional microscopic tissue structure characteristics.

[0031] In step A1, the surface morphology and internal structure of the inkjet-printed thin film sensor are observed by SEM and FIB to obtain a variety of microstructure maps; It should be noted that the scanning electron microscope (SEM) is an observation method between the transmission electron microscope and the optical microscope. It uses a focused, narrow high-energy electron beam to scan the sample, and stimulates various physical information through the interaction between the beam and the material. This information is collected, amplified, and re-imaged to achieve the purpose of characterizing the microscopic morphology of the material. The resolution of the new scanning electron microscope can reach 1nm; the magnification can reach 300,000 times or more and is continuously adjustable; and it has a large depth of field, a large field of view, and a good three-dimensional imaging effect. In addition, the combination of scanning electron microscopes and other analytical instruments can observe the microscopic morphology while analyzing the micro-area composition of the material. Scanning electron microscopes are widely used in the research of rock, soil, graphite, ceramics, and nanomaterials. Therefore, scanning electron microscopes play an important role in the field of scientific research; Focused ion beam (FIB) technology is a microdissection technology that uses an electrostatic lens to focus an ion beam into a very small size. The particle beam of a commercial FIB system is mostly derived from a liquid metal ion source. Since gallium has a low melting point, low vapor pressure, and good antioxidant properties, the metal material in the liquid metal ion source is mostly gallium (Ga). Applying an electric field (Suppressor) to the liquid metal ion source at the top of the ion column can form a small tip of liquid metal or alloy, and then a negative electric field (Extractor) pulls the metal or alloy at the tip to derive the ion beam, which is then focused by an electrostatic lens. After a series of variable apertures (Automatic Variable Aperture, AVA), the size of the ion beam can be determined, and then the required ion species are screened out by an E×B mass analyzer. Finally, the ion beam is focused on the sample and scanned by an octopole deflection device and an objective lens. The ion beam bombards the sample, and the secondary electrons and ions generated are collected and imaged or cut or polished by physical collision.

[0032] In step A2, the above-mentioned mesoscopic tissue map is converted into a binary map using an adaptive threshold algorithm, and on this basis, information such as the shape and area of ​​the voids in the map is statistically analyzed; The pseudo code of the adaptive threshold algorithm is: I1=imread(strcat(p,f)); % Input original image PS=5.5; %The actual width of each pixel value, unit: nm I1=I1(:,:,1); level = graythresh(I1); I1=im2bw(I1,level); The threshold is calculated by the code based on the microscopic tissue map.

[0033] In step A3, based on the stereological method, the two-dimensional tissue atlas information obtained in the above steps is converted into three-dimensional structural features to obtain microscopic structural feature information including void geometry, void ratio, etc.; It should be noted that stereology is a mathematical method that uses two-dimensional cross-sections or projection images to obtain three-dimensional structural information. It uses rigorous mathematical methods to obtain information from cross-sections that are smaller than the actual tissue dimension (such as two-dimensional cross-sections or one-dimensional cross-sections) to quantitatively describe the three-dimensional parameters of the actual tissue.

[0034] Preferably, a preset mesoscopic mechanical model and phase field fracture model of an inkjet-printed thin film sensor are obtained, specifically: a mesoscopic mechanical model of an inkjet-printed thin film sensor is constructed in abaqus finite element software according to Python script programming; The phase field fracture model is constructed using the uel program in the abaqus finite element software.

[0035] In the actual application process, the micromechanical model of the inkjet printed thin film sensor is constructed in the abaqus finite element software through Python script programming, and the phase field fracture model is constructed through the uel program in the abaqus finite element software; ABAQUS is a powerful finite element software for engineering simulation, which can solve problems ranging from relatively simple linear analysis to many complex nonlinear problems. ABAQUS includes a rich unit library that can simulate arbitrary geometric shapes. It also has various types of material model libraries that can simulate the performance of typical engineering materials, including metals, rubbers, polymer materials, composite materials, reinforced concrete, compressible hyperelastic foam materials, and geological materials such as soil and rock. As a general simulation tool, ABAQUS can not only solve a large number of structural (stress / displacement) problems, but also simulate many problems in other engineering fields, such as heat conduction, mass diffusion, thermoelectric coupling analysis, acoustic analysis, geotechnical mechanics analysis (fluid penetration / stress coupling analysis) and piezoelectric medium analysis.

[0036] like Figure 3 As shown, preferably, step S4 includes the following steps: B1. Input the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix into the micromechanical model of the inkjet printed thin film sensor to obtain the stress value corresponding to the inkjet process parameter matrix; B2. Input the stress value corresponding to the inkjet process parameter matrix into the phase field fracture model, obtain the moving distance of the indenter and the vertical load change, and integrate them to obtain the load displacement curve of the scratching head; B3. Determine whether the load displacement curve of the scratching head meets the preset conditions. If so, traverse the inkjet process parameter matrix to calculate the interface bonding strength as the real interface bonding strength.

[0037] From step B1 to step B3, the micromechanical model of the inkjet-printed thin film sensor is to calculate the corresponding stress value based on the three-dimensional microstructure characteristics; then the stress value is input into the phase field fracture model, which is written by the UEL subroutine in Abaqus. The crack extension is controlled by setting the stress. When the stress reaches the set initial value, the crack extends. When the crack extends to the film and the substrate, the film is considered to have failed. At this time, the distance moved by the scratch head and the change in the vertical load are measured. This is the load displacement curve image. The load displacement curve is compared with the pre-stored image in the database. If the comparison is consistent, the inkjet process parameter matrix is ​​traversed, and the calculated interface bonding strength is used to represent the real interface bonding strength. If it is inconsistent, return to the LSTM model training step and retrain the LSTM model.

[0038] Preferably, the process parameter-microstructure-interface bonding strength coupling structure-activity relationship is specifically: There is a mapping relationship between the inkjet process parameter matrix and the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix; The relationship between the three-dimensional microstructure characteristics and the interface bonding strength corresponding to the inkjet process parameter matrix is ​​a power function, and the specific formula is: ; in, is the interface bonding strength, It is the three-dimensional microstructure feature. and is the weight coefficient.

[0039] In actual application, the inkjet process parameter matrix is ​​formed by the combination of different process parameters under multi-dimensional process types. The inkjet process parameter matrix and the corresponding three-dimensional micro-structure characteristics are obtained through step S2. Therefore, the inkjet process parameter matrix and the corresponding three-dimensional micro-structure characteristics are a mapping relationship; the three-dimensional micro-structure characteristics corresponding to the inkjet process parameter matrix and the interface bonding strength are in a power function relationship, and the weight coefficients a and b need to be confirmed through fitting.

[0040] In this embodiment, the machine learning method is used to calculate the surface porosity (x) of the microstructure map of the ITO film with different process parameters, and the scratch test is used to calculate the interface bonding strength of different process parameters. The fitting parameters a and b are calculated using this formula.

[0041] like Figure 4 As shown, preferably, step S6 is specifically: C1. Determine the optimal process parameters in the inkjet process parameter matrix according to the genetic algorithm; C2. Determine the optimized interface bonding strength according to the optimal process parameters, and update the current interface strength of the inkjet printed thin film sensor with the optimized interface bonding strength.

[0042] In step C1 to step C2, the optimal process parameters are determined in the inkjet process parameter matrix through a genetic algorithm, and the optimized interface bonding strength is obtained through the coupled structure-activity relationship of process parameters-microstructure-interface bonding strength. The optimized interface strength is used to update the current interface strength of the ink-printed thin film sensor.

[0043] like Figure 5 As shown, an inkjet printed thin film sensor interface strength optimization system comprises: A three-dimensional microstructure feature module is used to observe and obtain the microstructure map of the inkjet-printed thin film sensor, and extract the three-dimensional microstructure features of the inkjet-printed thin film sensor after processing; A neural network training module is used to form an inkjet process parameter matrix formed by combining different process parameters under multi-dimensional process types as input variables, and to train the LSTM neural network with the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix as the target to obtain a trained neural network model; The real interface bonding strength module is used to calculate and obtain the load displacement curve of the scratching head according to the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix, the preset inkjet printing thin film sensor micromechanics model and the phase field fracture model, and obtain the real interface bonding strength according to the load displacement curve of the scratching head; The structure-activity relationship module is used to take the inkjet process parameter matrix as input variables and the three-dimensional microstructure characteristics and interface binding strength corresponding to the inkjet process parameter matrix as outputs, and to train the trained neural network model again to obtain the process parameter-microstructure-interface binding strength coupled structure-activity relationship; The interface bonding strength optimization module optimizes the current interface bonding strength based on the process parameters-microstructure-interface bonding strength coupling structure-activity relationship and genetic algorithm, and the corresponding process parameters in the inkjet process parameter matrix.

[0044] During actual application, the inkjet-printed thin film sensor interface strength optimization system is also equipped with a three-dimensional microstructure feature module, a neural network training module, a real interface binding strength module, a structure-activity relationship module and an interface binding strength optimization module. The functions of each module in the system correspond to the various steps in the implementation method and will not be repeated here.

[0045] In the embodiments provided in the present application, it should be understood that the disclosed methods and systems can be implemented in other ways. The system embodiments described above are merely schematic. For example, the division of modules is only a logical function division. There may be other division methods in actual implementation, such as: multiple modules or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0046] In addition, all functional modules in the embodiments of the present invention may be integrated into one processor, or each module may be a separate device, or two or more modules may be integrated into one device; each functional module in the embodiments of the present invention may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0047] A person of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by program instructions and related hardware. The aforementioned program instructions can be stored in a computer-readable storage medium. When the program instructions are executed, the steps of the above method embodiment are executed; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a magnetic disk or an optical disk, and other media that can store program codes.

[0048] It should be understood that the use of "system", "device", "unit" and / or "module" in this application is only a method for distinguishing different components, elements, parts, parts or assemblies at different levels. However, if other words can achieve the same purpose, the word can be replaced by other expressions.

[0049] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "a kind" and / or "the" do not refer to the singular, but also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of clearly identified steps and elements, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements. The elements defined by the sentence "includes a..." do not exclude the existence of other identical elements in the process, method, commodity or device that includes the elements.

[0050] If a flow chart is used in the present application, the flow chart is used to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, each step can be processed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or a certain step or several steps of operations can be removed from these processes.

[0051] The above is a detailed introduction to an inkjet-printed thin film sensor interface strength optimization method and system provided by the present invention. The above description of the disclosed embodiments enables professionals and technicians in this field to implement or use the present invention. Various modifications to these embodiments will be obvious to professionals and technicians in this field, and the general principles defined in this article can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown in this article, but will conform to the widest range consistent with the principles and novel features disclosed in this article.

Claims

1. A method for optimizing the interface strength of an inkjet printed thin film sensor, characterized in that: The steps include: Provide inkjet printed thin film sensors with different process parameters under multi-dimensional process types; Observe and obtain the microstructure map of the inkjet printed thin film sensor, and extract the three-dimensional microstructure characteristics of the inkjet printed thin film sensor after processing; An inkjet process parameter matrix formed by combining different process parameters under multi-dimensional process types is used as an input variable, and a three-dimensional microstructure feature corresponding to the inkjet process parameter matrix is ​​used as a target to train the LSTM neural network to obtain a trained neural network model; According to the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix, the preset inkjet printing thin film sensor micromechanics model and the phase field fracture model, the load displacement curve of the scratching head is calculated and obtained, and the real interface bonding strength is obtained according to the load displacement curve of the scratching head; Taking the inkjet process parameter matrix as input variables, taking the three-dimensional microstructure characteristics and the interface bonding strength corresponding to the inkjet process parameter matrix as outputs, the trained neural network model is trained again to obtain the process parameter-microstructure-interface bonding strength coupled structure-activity relationship; According to the process parameter-microstructure-interface bonding strength coupled structure-activity relationship and genetic algorithm, the corresponding process parameters are searched in the inkjet process parameter matrix to optimize the current interface bonding strength.

2. The method for optimizing the interface strength of an inkjet printed thin film sensor according to claim 1, characterized in that: The method of observing and obtaining the microstructure map of the inkjet printed thin film sensor and extracting the three-dimensional microstructure characteristics of the inkjet printed thin film sensor after processing comprises the following steps: Observe the inkjet-printed thin film sensors with different process parameters under various multi-dimensional process types to obtain corresponding various micro-structure maps; According to an adaptive threshold algorithm, the mesoscopic tissue structure map is converted into a binary tissue map; Based on stereological methods, the tissue atlas is converted into three-dimensional microscopic tissue structural features.

3. The inkjet printed thin film sensor interface strength optimization method according to claim 1, characterized in that: Obtaining a preset mesoscopic mechanical model and phase field fracture model of an inkjet-printed thin film sensor, specifically: constructing the mesoscopic mechanical model of the inkjet-printed thin film sensor in abaqus finite element software according to Python script programming; The phase field fracture model is constructed using the uel program in the abaqus finite element software.

4. The method for optimizing the interface strength of an inkjet printed thin film sensor according to claim 3, characterized in that: The method comprises the following steps: calculating and obtaining a load displacement curve of a scratching head according to the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix, a preset micromechanical model of an inkjet printed thin film sensor, and a phase field fracture model, and obtaining a true interface bonding strength according to the load displacement curve of the scratching head: Inputting the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix into the micromechanical model of the inkjet printed thin film sensor to obtain the stress value corresponding to the inkjet process parameter matrix; Inputting the stress value corresponding to the inkjet process parameter matrix into the phase field fracture model, obtaining the moving distance of the indenter and the vertical load change, and integrating to obtain the load displacement curve of the scratching head; It is determined whether the load displacement curve of the scratching head meets a preset condition. If so, the inkjet process parameter matrix is ​​traversed to calculate the interface bonding strength as the real interface bonding strength.

5. The method for optimizing the interface strength of an inkjet printed thin film sensor according to claim 1, characterized in that: The process parameter-microstructure-interface bonding strength coupling structure-activity relationship is specifically: There is a mapping relationship between the inkjet process parameter matrix and the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix; The three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix and the interface bonding strength are in a power function relationship, and the specific formula is: ； in, is the interface bonding strength, It is the three-dimensional microstructure feature. and is the weight coefficient.

6. The method for optimizing the interface strength of an inkjet printed thin film sensor according to claim 1, characterized in that: According to the process parameter-microstructure-interface bonding strength coupling structure-activity relationship and genetic algorithm, the corresponding process parameters in the inkjet process parameter matrix are used to optimize the current interface bonding strength, specifically: Determining optimal process parameters in the inkjet process parameter matrix according to the genetic algorithm; The optimized interface bonding strength is determined according to the optimal process parameters, and the current interface strength of the inkjet printed thin film sensor is updated with the optimized interface bonding strength.

7. An inkjet printed thin film sensor interface strength optimization system, characterized in that: include: A three-dimensional microstructure feature module is used to observe and obtain the microstructure map of the inkjet-printed thin film sensor, and extract the three-dimensional microstructure features of the inkjet-printed thin film sensor after processing; A neural network training module is used to form an inkjet process parameter matrix formed by combining different process parameters under multi-dimensional process types as input variables, and to train the LSTM neural network with the three-dimensional microstructure characteristics corresponding to the inkjet process parameter matrix as a target to obtain a trained neural network model; A real interface bonding strength module is used to calculate and obtain the load displacement curve of the scratching head according to the three-dimensional microscopic structural characteristics corresponding to the inkjet process parameter matrix, the preset inkjet printing thin film sensor microscopic mechanical model and the phase field fracture model, and obtain the real interface bonding strength according to the load displacement curve of the scratching head; A structure-activity relationship module is used to use the inkjet process parameter matrix as an input variable, and the three-dimensional microstructure characteristics and the interface binding strength corresponding to the inkjet process parameter matrix as outputs, and to train the trained neural network model again to obtain the process parameter-microstructure-interface binding strength coupled structure-activity relationship; The interface bonding strength optimization module optimizes the current interface bonding strength according to the process parameter-microstructure-interface bonding strength coupling structure-activity relationship and the genetic algorithm, and the corresponding process parameters in the inkjet process parameter matrix.

Images