A conduction loss analysis method and system for flexible device modules

Abstract

The present invention relates to the field of information technology, and discloses a conduction loss analysis method and system for flexible device modules. The screening range of flexible substrate materials is determined by performance parameters of a target flexible device module, and the correlation between deformation evolution curves and mechanical and electrical characteristics is established according to the stress and resistance changes of the flexible device materials to be determined within the range under various loads. The two are combined to form a performance index matrix for preliminary screening for screening. A characteristic matrix reflecting the change of material performance over time is constructed using aging experimental data of the materials after preliminary screening, and a time domain curve of conduction loss is further obtained. The performance of primary materials is quantitatively scored through a multi-objective genetic optimization function to select the optimal target material. Based on the conduction loss data of the target material, and in combination with the mechanical and electrical characteristics of the flexible substrate material and the performance changes during the aging process, an accurate analysis of the conduction loss of the flexible device module is achieved.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a conduction loss analysis method and system for a flexible device module. Background Art

[0002] In the design and manufacturing process of flexible device modules, conduction loss is a key technical issue. Different flexible substrate materials have different mechanical and electrical properties, such as Young's modulus, Poisson's ratio, resistivity parameters, etc. The differences in these parameters will cause the flexible device module to deform to varying degrees under different working conditions due to different stress states, thereby affecting the resistance distribution and conduction loss inside the module. At present, circuit simulation software is usually used to simulate the current distribution and conduction loss in flexible device modules, but flexible device modules are prone to deformation in actual applications, resulting in increased conduction loss. Due to the limitations of the simulation environment, circuit simulation is difficult to simulate the phenomenon of increased conduction loss caused by deformation, resulting in low accuracy of the analysis results.

[0003] It can be seen that how to solve the problem of insufficient accuracy of existing conduction loss analysis methods has become a technical problem that technical personnel in this field need to solve urgently. Summary of the invention

[0004] The present invention provides a conduction loss analysis method and system for a flexible device module, which solves the problem of how to improve the accuracy of the existing conduction loss analysis method.

[0005] In order to solve the above technical problems, the first aspect of the present invention provides a conduction loss analysis method for a flexible device module, comprising:

[0006] Acquiring performance parameters of a target flexible device module, and selecting a to-be-determined flexible substrate material group from a plurality of flexible substrate materials according to the performance parameters;

[0007] According to the force changes and resistance changes of the to-be-determined flexible substrate material group under various load conditions, a mapping relationship between the mechanical characteristics and the electrical characteristics of the to-be-determined flexible substrate material group is constructed;

[0008] Constructing a deformation evolution curve based on deformation data of the undetermined flexible substrate material group under several curvature radii, and generating a performance index matrix in combination with the mapping relationship, so as to select a primary flexible substrate material from the undetermined flexible substrate material group;

[0009] Constructing a characteristic mapping matrix according to the time-domain deformation data and the time-domain resistance change data of the primary flexible substrate material under multiple aging experiments, and constructing a conduction loss time-domain curve of the primary flexible substrate material during the aging process according to the characteristic mapping matrix;

[0010] Constructing a multi-objective genetic optimization function according to the conduction loss time domain curve, and quantifying the material properties of the primary flexible substrate material based on the multi-objective genetic optimization function, and selecting a target flexible substrate material from the primary flexible substrate material according to the quantification result;

[0011] The conduction loss data of the target flexible substrate material is quantified, and the conduction loss value of the target flexible device module is determined according to the conduction loss data, so as to realize the conduction loss analysis of the target flexible device module.

[0012] As one preferred solution, the selecting a to-be-determined flexible substrate material group from a plurality of flexible substrate materials according to the performance parameters comprises:

[0013] Determine the electrical conductivity threshold, thermal conductivity threshold, stress threshold, strain threshold, interlayer bonding strength threshold and fatigue resistance threshold of the flexible substrate material according to the performance parameters;

[0014] Acquiring the electrical conductivity and thermal conductivity of the plurality of flexible substrate materials, and selecting the flexible substrate materials whose electrical conductivity exceeds the electrical conductivity threshold and whose corresponding thermal conductivity exceeds the thermal conductivity threshold as a pre-selected material group;

[0015] Calculating stress data and strain data of the preselected material group under a preset external force by a finite element analysis method, and taking the preselected material group whose stress data exceeds the stress threshold and whose corresponding strain data exceeds the strain threshold as a candidate material group;

[0016] Obtaining interlayer bonding strength data of the candidate material group and fatigue resistance values ​​obtained by performing a cyclic bending test on the candidate material group, and selecting a candidate material group whose interlayer bonding strength data exceeds the interlayer bonding strength threshold and whose corresponding fatigue resistance values ​​exceed the fatigue resistance threshold as a candidate material group;

[0017] The adaptability score of the candidate material group is quantified under the coupling of thermal stress and mechanical stress, and the candidate material group that meets the adaptability requirement is used as the to-be-determined flexible substrate material group.

[0018] As one preferred solution, quantifying the adaptability score of the candidate material group under the coupling of thermal stress and mechanical stress includes:

[0019] Obtaining the thermal expansion coefficient, strain allowable range and Young's modulus of the candidate material group to quantify the stress state of the candidate material group under the coupling of thermal stress and mechanical stress, and obtaining mechanical stress and thermal stress;

[0020] The mechanical stress and the thermal stress are combined by finite element analysis to obtain a coupled stress field, and the safety margin, stress concentration factor and fatigue life of the alternative material group are evaluated based on the coupled stress field to quantify the adaptability score of the alternative material group by a weighted average method.

[0021] As one of the preferred solutions, constructing a mapping relationship between the mechanical characteristics and the electrical characteristics of the to-be-determined flexible substrate material group according to the force changes and resistance changes of the to-be-determined flexible substrate material group under various load conditions includes:

[0022] The mechanical parameters of the to-be-determined flexible substrate material group are collected to construct a tetrahedral unit space grid, and the grid cells in the tetrahedral unit space grid are refined by boundary point constraints to obtain a tetrahedral grid calculation graph;

[0023] Obtaining a displacement cloud map and a strain field distribution map after applying multiple load conditions to the tetrahedral grid calculation map, and extracting the displacement and strain distribution data of the undetermined flexible substrate material group after deformation based on the displacement cloud map and the strain field distribution map, to obtain a stress-strain curve of the undetermined flexible substrate material group;

[0024] Substituting the displacement and the strain distribution data into the Pearson resistance-strain relationship formula to calculate the conductivity gradient of the undetermined flexible substrate material group, and spatially interpolating the conductivity gradient through a deep neural network to obtain conductivity distribution data to construct a local load compensation matrix;

[0025] Based on the local load compensation matrix, a support vector regressor is used to calculate the resistance change of the undetermined flexible substrate material group under various load conditions to obtain a resistance change curve, and a mapping relationship between the mechanical characteristics and the electrical characteristics of the undetermined flexible substrate material group is constructed according to the resistance change curve and the stress-strain curve.

[0026] As one of the preferred solutions, the mapping relationship between the mechanical characteristics and the electrical characteristics of the to-be-determined flexible substrate material group is constructed according to the resistance change curve and the stress-strain curve, including:

[0027] Fitting the stress-strain curve by the least square method, and determining the turning point position of the elastic deformation and plastic deformation of the flexible substrate material group to be determined by the first-order derivative change rate of the fitting curve, so as to obtain standardized mechanical characteristic parameters;

[0028] Based on the resistance change curve, the resistance change rate is calculated by sliding the time window, and the time domain characteristic value of the resistance change rate is collected to obtain the standardized electrical characteristic parameters of the to-be-determined flexible substrate material group;

[0029] The standardized mechanical characteristic parameters and the standardized electrical characteristic parameters are used as sample data to train a neural network model, and the associated weights of stress strain and resistance change are added between the hidden layer nodes of the neural network model, and the weight gradient is calculated by back propagation to obtain a trained neural network model;

[0030] The parameters of the trained neural network model are optimized through a radial basis function support vector machine regressor, and the optimal parameter combination is selected by cross-validation. When the prediction error of the neural network model after parameter adjustment reaches the error accuracy, the mapping relationship between the mechanical characteristics and electrical characteristics of the undetermined flexible substrate material group is obtained.

[0031] As one of the preferred solutions, the deformation evolution curve is constructed based on the deformation data of the undetermined flexible substrate material group under several curvature radii, and a performance index matrix is ​​generated in combination with the mapping relationship to select a primary flexible substrate material from the undetermined flexible substrate material group, including:

[0032] Obtaining deformation data of the undetermined flexible substrate material group under several curvature radii, obtaining a curvature displacement distribution cloud map to train a convolutional neural network as sample data, obtaining a deformation prediction model to output a predicted deformation amount of the undetermined flexible substrate material group under several cyclic bending conditions, and constructing the deformation evolution curve based on the predicted deformation amount;

[0033] Obtaining resistance change rate data of the to-be-determined flexible substrate material group at several curvature radii, and constructing a fifth-order polynomial regressor based on the mapping relationship to calculate the quantitative correspondence between the resistance change rate data and the data in the deformation evolution curve;

[0034] According to the quantitative correspondence, a performance index matrix is ​​constructed, wherein the row vectors are the radius of curvature and the number of bending times, and the column vectors are the deformation amount and the resistance change amount. By performing weighted calculation on the performance index matrix, the comprehensive score of the undetermined flexible substrate material group is obtained and sorted to select primary flexible substrate materials in order.

[0035] As one of the preferred solutions, constructing a feature mapping matrix according to the time-domain deformation data and the time-domain resistance change data of the primary flexible substrate material under multiple aging experiments, and constructing a conduction loss time-domain curve of the primary flexible substrate material during the aging process according to the feature mapping matrix, includes:

[0036] Acquire the time-domain deformation data of the primary flexible substrate material under multiple aging experiments to construct a deformation variable time-domain curve, and use a Wheatstone bridge to measure the time-domain resistance change data based on the deformation variable time-domain curve to construct a resistance change time-domain curve, and use the deformation variable time-domain curve as sample data to train a deep neural network to obtain the feature mapping matrix;

[0037] Based on the characteristic mapping matrix, the conduction loss value of the primary flexible substrate material is quantified by the Joule heat calculation formula to construct a conduction loss change curve, and the conduction loss change curve is decomposed on different time scales by using a three-layer wavelet transform to obtain three components of long-term trend, periodic fluctuation and random disturbance;

[0038] The prediction time span and confidence threshold are set, and the components are quantized by an autoregressor to obtain a time domain curve of the conduction loss of the primary flexible substrate material during the aging process.

[0039] As one of the preferred solutions, constructing a multi-objective genetic optimization function according to the conduction loss time domain curve, quantifying the material properties of the primary flexible substrate material based on the multi-objective genetic optimization function, and selecting a target flexible substrate material from the primary flexible substrate material according to the quantification result, comprises:

[0040] The conduction loss time domain curve is subjected to piecewise linearization and standardization processing to obtain a piecewise standardized conduction loss function to construct the multi-objective genetic optimization function, and the parameters of the multi-objective genetic optimization function are optimized by genetic crossover and mutation operations to obtain a target conduction loss curve;

[0041] quantifying the root mean square error between the conduction loss time domain curve of the primary flexible substrate material and the target conduction loss curve by the least square method, and fitting the root mean square error by a radial basis kernel function support vector machine to quantify the material performance of the primary flexible substrate material and obtain a material performance score;

[0042] A screening matrix is ​​established according to the material performance score, and the loss growth rate and the loss baseline value are weighted by the screening matrix to calculate the weighted score of the primary flexible substrate material, and the primary flexible substrate material whose weighted score meets the weighted score requirement is used as the target flexible substrate material.

[0043] As one preferred solution, the quantifying the conduction loss data of the target flexible substrate material includes:

[0044] Acquiring temperature field data of the target flexible substrate material to construct a temperature field distribution cloud map, and constructing a finite element calculation grid based on the temperature field distribution cloud map, so as to obtain the thermal conductivity and thermal diffusion coefficient of the target flexible substrate material through heat flux density calculation;

[0045] The thermal conductivity coefficient and the thermal diffusion coefficient are used as sample data to train a three-layer hidden layer neural network to obtain a thermal conductivity prediction model to output temperature distribution data, and according to the temperature distribution data, a power loss calculation formula based on Joule's heat law is used to quantify the conduction loss rate at each sampling point, so as to fit the temperature distribution data and the conduction loss rate through a third-order polynomial fitter to obtain a conduction loss distribution diagram;

[0046] The long short-term memory network is trained through the conduction loss distribution map to obtain a conduction loss optimization prediction model to output the power consumption change prediction trend, and when the power consumption change prediction trend meets the design index requirements of the target flexible device module, the power consumption change prediction trend is output as the conduction loss data of the target flexible substrate material.

[0047] A second aspect of the present invention provides a conduction loss analysis system for a flexible device module, comprising:

[0048] A range determination module, used for obtaining performance parameters of a target flexible device module, and selecting a to-be-determined flexible substrate material group from a plurality of flexible substrate materials according to the performance parameters;

[0049] A relationship building module, used to build a mapping relationship between the mechanical characteristics and electrical characteristics of the to-be-determined flexible substrate material group according to the force changes and resistance changes of the to-be-determined flexible substrate material group under various load conditions;

[0050] A primary screening module, for constructing a deformation evolution curve based on deformation data of the undetermined flexible substrate material group under several curvature radii, and generating a performance index matrix in combination with the mapping relationship, so as to select a primary flexible substrate material from the undetermined flexible substrate material group;

[0051] A curve construction module, used to construct a feature mapping matrix according to the time-domain deformation data and the time-domain resistance change data of the primary flexible substrate material under multiple aging experiments, and to construct a conduction loss time-domain curve of the primary flexible substrate material during the aging process according to the feature mapping matrix;

[0052] a target selection module, configured to construct a multi-objective genetic optimization function according to the conduction loss time domain curve, quantify the material properties of the primary flexible substrate material based on the multi-objective genetic optimization function, and select a target flexible substrate material from the primary flexible substrate material according to the quantification result;

[0053] The loss analysis module is used to quantify the conduction loss data of the target flexible substrate material and determine the conduction loss value of the target flexible device module according to the conduction loss data to implement conduction loss analysis of the target flexible device module.

[0054] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0055] (1) The mechanical and electrical properties of flexible substrate materials and the performance changes during aging are comprehensively considered. By constructing a multi-dimensional mapping relationship and indicator matrix, a comprehensive evaluation of material performance is achieved to improve the accuracy of conduction loss evaluation of the target flexible device module;

[0056] (2) Through multiple aging experiments and the collection and analysis of deformation data and time domain data, it can accurately reflect the changing trend of material properties over time, providing a reliable basis for screening target materials with excellent performance;

[0057] (3) The material properties are quantitatively scored using a multi-objective genetic optimization function, which greatly improves the screening efficiency. At the same time, the conduction loss analysis of the target flexible device module also has strong practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1 It is a flow chart of a conduction loss analysis method for a flexible device module provided by a certain embodiment of the present invention;

[0060] Figure 2 It is a structural diagram of a conduction loss analysis system for a flexible device module provided by a certain embodiment of the present invention. DETAILED DESCRIPTION

[0061] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings and embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. 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.

[0062] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0063] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two components. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the system or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0064] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0065] In one embodiment, if Figure 1 As shown, the first aspect of the present invention provides a conduction loss analysis method for a flexible device module, comprising:

[0066] S1. Obtaining performance parameters of a target flexible device module, and selecting a to-be-determined flexible substrate material group from a plurality of flexible substrate materials according to the performance parameters;

[0067] Specifically, the present invention takes into account that the existing circuit simulation software cannot consider the influence of different types of flexible substrate materials on the conduction loss of the flexible device module composed of it when simulating the current distribution in the flexible device module to perform the conduction loss method, and the problem that the flexible device module is prone to deformation in actual application, resulting in increased conduction loss. The target flexible device module is split and the target flexible substrate material is selected according to the performance of the device module, so as to determine the conduction loss of the target flexible device module based on the conduction loss data of the target flexible substrate material after comprehensive consideration of the mechanical properties, electrical properties, deformation process and aging process, so as to improve its accuracy; wherein, the number and type of the target flexible device module are not specifically limited here, and its performance parameters include mechanical strength, electrical connection and thermal management and other related parameters used to characterize the performance of the target flexible device module, so as to determine the selectable range of the flexible substrate material according to the performance parameters. It should be noted that the target flexible device module finally selected needs to be connected with the corresponding electronic components and connectors to form the target flexible device module.

[0068] In one embodiment, selecting a to-be-determined flexible substrate material group from a plurality of flexible substrate materials according to the performance parameter comprises:

[0069] Determine the electrical conductivity threshold, thermal conductivity threshold, stress threshold, strain threshold, interlayer bonding strength threshold and fatigue resistance threshold of the flexible substrate material according to the performance parameters;

[0070] Acquiring the electrical conductivity and thermal conductivity of the plurality of flexible substrate materials, and selecting the flexible substrate materials whose electrical conductivity exceeds the electrical conductivity threshold and whose corresponding thermal conductivity exceeds the thermal conductivity threshold as a pre-selected material group;

[0071] Calculating stress data and strain data of the preselected material group under a preset external force by a finite element analysis method, and taking the preselected material group whose stress data exceeds the stress threshold and whose corresponding strain data exceeds the strain threshold as a candidate material group;

[0072] Obtaining interlayer bonding strength data of the candidate material group and fatigue resistance values ​​obtained by performing a cyclic bending test on the candidate material group, and selecting a candidate material group whose interlayer bonding strength data exceeds the interlayer bonding strength threshold and whose corresponding fatigue resistance values ​​exceed the fatigue resistance threshold as a candidate material group;

[0073] The adaptability score of the candidate material group is quantified under the coupling of thermal stress and mechanical stress, and the candidate material group that meets the adaptability requirement is used as the to-be-determined flexible substrate material group.

[0074] Specifically, the present invention determines the upper limit of the parameters of the flexible substrate material according to the performance parameters of the target flexible device module, including electrical conductivity, thermal conductivity, stress, strain, interlayer bonding strength and fatigue resistance, and then determines the undetermined flexible substrate material group in the selectable range of the flexible substrate material based on these parameters; wherein each undetermined flexible substrate material group is a basic material group that can be connected with the corresponding electronic components and connectors to form the target flexible device module, specifically:

[0075] Flexible substrate materials with electrical conductivity greater than 0.1 Siemens per meter and thermal conductivity greater than 200 watts per meter Kelvin were selected as the pre-selected material group; finite element software was used to calculate the stress-strain distribution cloud map of the pre-selected material group under the action of 300 Newton external force, and the pre-selected material group with stress less than 200 MPa and strain less than 5% was selected as the candidate material group; the fatigue resistance value of the candidate material group under the condition of 6000 bending radius of 3 mm was measured by a cyclic bending tester, and the flexible substrate materials with interlayer bonding strength greater than 4 MPa and fatigue resistance index greater than 80 points were selected as the alternative material group; for the alternative material group, a strain gauge tester was selected to measure the allowable strain value, a dynamic elastic modulus meter was used to measure the Young's modulus value, and a random forest regressor was used to construct a comprehensive scoring model to calculate the strain adaptability score under the coupling of thermal stress and mechanical stress, and the alternative material group with a strain adaptability score greater than 90 points and a Young's modulus less than 3 GPa was selected as the undetermined flexible substrate material group.

[0076] The present invention evaluates the key performance indicators of the flexible substrate material to ensure that the selected material meets the specific application requirements; uses the finite element analysis method to calculate stress data and strain data, thereby improving the calculation accuracy and prediction ability, and helping to more accurately evaluate the performance of the material under the action of external forces; comprehensively evaluates the flexible substrate material from multiple dimensions such as electrical conductivity, thermal conductivity, mechanical properties (stress and strain), interlayer bonding strength, and fatigue resistance to ensure that the selected material has excellent performance in all aspects; and forms a systematic screening process through step-by-step screening (pre-selected material group → candidate material group → alternative material group), which helps to gradually narrow the selection range and improve screening efficiency.

[0077] In one embodiment, quantifying the adaptability score of the candidate material group under the coupling of thermal stress and mechanical stress includes:

[0078] Obtaining the thermal expansion coefficient, strain allowable range and Young's modulus of the candidate material group to quantify the stress state of the candidate material group under the coupling of thermal stress and mechanical stress, and obtaining mechanical stress and thermal stress;

[0079] The mechanical stress and the thermal stress are combined by finite element analysis to obtain a coupled stress field, and the safety margin, stress concentration factor and fatigue life of the alternative material group are evaluated based on the coupled stress field to quantify the adaptability score of the alternative material group by a weighted average method.

[0080] Specifically, the present invention quantifies the adaptability score of the candidate material group under the action of thermomechanical coupling: first, the thermal expansion coefficient, strain allowable range and Young's modulus of the candidate material group are obtained, and these data are used to determine the mechanical stress and thermal stress of the candidate material group; wherein, the mechanical stress is the stress caused by the external load, obtained by mechanical analysis, and the thermal stress is the stress caused by the temperature change, obtained by thermoelastic mechanical analysis; then, the temperature field and stress field are solved by the finite element analysis method to combine the mechanical stress and thermal stress to calculate the coupled stress field, and other numerical analysis methods can also be used to accurately simulate the complex stress state of the candidate material group, and evaluate the adaptability of the candidate material group according to the coupled stress field, that is, the safety margin, stress concentration factor and fatigue life; finally, different weights are assigned to each evaluation index according to its importance, and the weighted average score is calculated to obtain the adaptability score of the candidate material group. In addition, for the calculation of the adaptability score of the candidate material group, the fuzzy comprehensive evaluation method can also be used to convert the evaluation index into a fuzzy set, and the adaptability score is obtained by fuzzy operation, and the multi-objective optimization method is used to regard the adaptability evaluation as a multi-objective optimization problem, and the adaptability score is obtained by solving the optimization model.

[0081] The present invention obtains key material parameters of the candidate material group to comprehensively quantify the stress state of the material under the coupling of thermal stress and mechanical stress; simulates complex stress fields through finite element analysis to improve the accuracy of evaluation; comprehensively evaluates material performance from key performance indicators such as material stress state, safety margin, stress concentration factor and fatigue life, which helps to have a more comprehensive understanding of the reliability and durability of the material; quantifies the adaptability score of the candidate material group through the weighted average method, which helps to find the most suitable material for the target flexible device module.

[0082] S2. constructing a mapping relationship between mechanical characteristics and electrical characteristics of the to-be-determined flexible substrate material group according to the force changes and resistance changes of the to-be-determined flexible substrate material group under various load conditions;

[0083] In one embodiment, step S2 includes:

[0084] The mechanical parameters of the to-be-determined flexible substrate material group are collected to construct a tetrahedral unit space grid, and the grid cells in the tetrahedral unit space grid are refined by boundary point constraints to obtain a tetrahedral grid calculation graph;

[0085] Obtaining a displacement cloud map and a strain field distribution map after applying multiple load conditions to the tetrahedral grid calculation map, and extracting the displacement and strain distribution data of the undetermined flexible substrate material group after deformation based on the displacement cloud map and the strain field distribution map, to obtain a stress-strain curve of the undetermined flexible substrate material group;

[0086] Substituting the displacement and the strain distribution data into the Pearson resistance-strain relationship formula to calculate the conductivity gradient of the undetermined flexible substrate material group, and spatially interpolating the conductivity gradient through a deep neural network to obtain conductivity distribution data to construct a local load compensation matrix;

[0087] Based on the local load compensation matrix, a support vector regressor is used to calculate the resistance change of the undetermined flexible substrate material group under various load conditions to obtain a resistance change curve, and a mapping relationship between the mechanical characteristics and the electrical characteristics of the undetermined flexible substrate material group is constructed according to the resistance change curve and the stress-strain curve.

[0088] Specifically, the present invention obtains Young's modulus, Poisson's ratio, and material density data of a to-be-determined flexible substrate material group to construct a tetrahedral unit space grid. In the process, each node has displacement degrees of freedom and rotation degrees of freedom in three directions of x, y, and z. Taking a polyimide flexible substrate as an example, the Young's modulus is 2.8 GPa, the Poisson's ratio is 0.34, and the material density is 1420 kg per cubic meter. A grid is generated on a substrate with a length of 200 mm, a width of 100 mm, and a thickness of 0.1 mm. The unit size is set to 0.1 mm, all degrees of freedom are restricted at the fixed end boundary, and deformation is allowed at the free end boundary. The distorted unit is removed through a grid quality check to obtain a tetrahedral unit; then, the grid units in the tetrahedral unit space grid are refined through fixed boundary point constraints and free boundary point constraints to obtain a tetrahedral grid calculation diagram in which each unit contains 6 degrees of freedom;

[0089] Apply three load conditions of stretching, bending and torsion to the tetrahedral grid calculation diagram, set the tensile load to 300 Newtons, the bending angle to 30 degrees, and the torsion moment to 2 Newton meters, and use the tetrahedral unit finite element stiffness matrix to calculate the stress distribution cloud map, output the displacement cloud map and strain field distribution map after the maximum stress value is less than the yield strength, and obtain the displacement and strain distribution data of the undetermined flexible substrate material group after deformation, so as to construct the stress-strain curve of the undetermined flexible substrate material group based on these data;

[0090] Then, the displacement and strain distribution data are substituted into the Pearson resistance-strain relationship formula, where the strain sensitivity coefficient is 2.1, the initial resistivity is 1.6×10^-8 ohm-meter, the resistance increases by 5% under tensile deformation, and returns to its original value after the load is removed. The conductivity matrix is ​​constructed through quadrilateral units, the conductivity gradient is calculated in each quadrilateral unit, and the conductivity gradient is spatially interpolated using a deep neural network to output the conductivity distribution data of each position point; the deep neural network adopts a 5-layer structure, the input layer receives the strain field data, the number of hidden layer nodes is 128, 256, 256, and 128 respectively, the output layer predicts the conductivity distribution, and the training data contains 1000 sets of strain and conductivity correspondence;

[0091] Finally, a local load compensation matrix is ​​constructed according to the conductivity distribution data, and the support vector regressor is used to calculate the local resistance change under three load conditions: tension, bending and torsion. The resistance change curve corresponding to different continuous loads and boundary constraints is obtained, and the mapping relationship between the mechanical characteristics and electrical characteristics of the undetermined flexible substrate material group is constructed according to the resistance change curve and the stress-strain curve. Among them, the resistance change compensation adopts a 3×3 local matrix, and the strain influence of the surrounding units is considered. The support vector regressor kernel function uses the radial basis function.

[0092] The present invention obtains a high-precision tetrahedral mesh calculation diagram through accurate simulation and detailed analysis of mechanical parameters, which helps to more accurately simulate the deformation and stress-strain distribution of materials under complex load conditions, thereby improving the accuracy of the analysis; not only the mechanical properties of the material are taken into account, but also the changes in its electrical properties are deeply explored, providing a basis for constructing a mapping relationship between mechanical characteristics and electrical characteristics; the combination of mechanical and electrical property analysis of the material has wide practicality and can be applied to various fields such as flexible electronics, smart materials, wearable devices, etc., to provide a scientific basis for material selection, design and optimization.

[0093] In one embodiment, the step of constructing a mapping relationship between the mechanical characteristics and the electrical characteristics of the to-be-determined flexible substrate material group according to the resistance change curve and the stress-strain curve includes:

[0094] Fitting the stress-strain curve by the least square method, and determining the turning point position of the elastic deformation and plastic deformation of the flexible substrate material group to be determined by the first-order derivative change rate of the fitting curve, so as to obtain standardized mechanical characteristic parameters;

[0095] Based on the resistance change curve, the resistance change rate is calculated by sliding the time window, and the time domain characteristic value of the resistance change rate is collected to obtain the standardized electrical characteristic parameters of the to-be-determined flexible substrate material group;

[0096] The standardized mechanical characteristic parameters and the standardized electrical characteristic parameters are used as sample data to train a neural network model, and the associated weights of stress strain and resistance change are added between the hidden layer nodes of the neural network model, and the weight gradient is calculated by back propagation to obtain a trained neural network model;

[0097] The parameters of the trained neural network model are optimized through a radial basis function support vector machine regressor, and the optimal parameter combination is selected by cross-validation. When the prediction error of the neural network model after parameter adjustment reaches the error accuracy, the mapping relationship between the mechanical characteristics and electrical characteristics of the undetermined flexible substrate material group is obtained.

[0098] Specifically, the present invention uses the least squares method to fit the stress-strain curve, calculates the elastic modulus in the stress range of less than 100 MPa, determines the yield strength value when the strain exceeds 0.2%, and judges the turning point position of elastic deformation and plastic deformation by the first-order derivative change rate of the fitting curve, and divides the elastic modulus measured at this time by the standard modulus to obtain the normalized elastic modulus, and divides the yield strength by the standard strength to obtain the normalized yield strength, and the two are the standardized mechanical characteristic parameters of the flexible substrate material group to be determined;

[0099] For the resistance change curve, the resistance change rate is calculated by sliding the time window, the window length is set to 100 milliseconds, the sliding step is 10 milliseconds, and the time domain characteristic value of the resistance change rate, that is, the resistivity, is collected, so that the resistance change rate and the resistivity are normalized based on the standard value of the resistance change rate and the standard value of the resistivity, respectively, to obtain the normalized resistance change rate and the normalized resistivity, and the standardized electrical characteristic parameters of the undetermined flexible substrate material group can be obtained;

[0100] Then, the training data matrix of the neural network model is constructed by using standardized mechanical characteristic parameters and standardized electrical characteristic parameters. The number of nodes in the input layer of the neural network model is set to 4, including four normalized parameters. The number of nodes in the two hidden layers is 16 and 8 respectively, and the number of nodes in the output layer is 2. The associated weights reflecting the stress strain and resistance change are added between the hidden layer nodes, and the weight gradient is calculated by back propagation. During the training process, the learning rate is set to 0.01, and the training error is calculated once every 100 iterations. When the error change is less than 0.001 for 10 consecutive times, the training is stopped, and the trained neural network model can be obtained. Among them, when the material undergoes elastic deformation, the resistance change is linearly related to the strain, and the weight value is small; after entering the plastic deformation, the resistance change is accelerated and the weight value increases;

[0101] Finally, the radial basis function support vector machine regressor is used to optimize the parameters of the trained neural network model. The penalty factor is set to 100, the radial basis function bandwidth is set to 0.1, and cross-validation is used to select the optimal parameter combination. When the prediction error of the neural network model after parameter adjustment reaches the error accuracy, that is, less than 0.01, the mapping relationship model is obtained to determine the mapping relationship between the yield strength and the resistance change rate of the undetermined flexible substrate material group.

[0102] The present invention performs precise data processing and analysis through the least squares method, sliding time window, etc., and can accurately describe the changes in the mechanical and electrical characteristics of the material; adding the associated weights of stress strain and resistance change between the hidden layer nodes of the neural network model can better capture the complex relationship between mechanical and electrical characteristics and improve the performance of the model; the precise construction of the mapping relationship is of great significance for understanding the mechanical and electrical behavior of the material and guiding the design and optimization of the material.

[0103] S3, constructing a deformation evolution curve based on the deformation data of the undetermined flexible substrate material group under several curvature radii, and generating a performance index matrix in combination with the mapping relationship, so as to select a primary flexible substrate material from the undetermined flexible substrate material group;

[0104] In one embodiment, step S3 includes:

[0105] Obtaining deformation data of the undetermined flexible substrate material group under several curvature radii, obtaining a curvature displacement distribution cloud map to train a convolutional neural network as sample data, obtaining a deformation prediction model to output a predicted deformation amount of the undetermined flexible substrate material group under several cyclic bending conditions, and constructing the deformation evolution curve based on the predicted deformation amount;

[0106] Obtaining resistance change rate data of the to-be-determined flexible substrate material group at several curvature radii, and constructing a fifth-order polynomial regressor based on the mapping relationship to calculate the quantitative correspondence between the resistance change rate data and the data in the deformation evolution curve;

[0107] According to the quantitative correspondence, a performance index matrix is ​​constructed, wherein the row vectors are the radius of curvature and the number of bending times, and the column vectors are the deformation amount and the resistance change amount. By performing weighted calculation on the performance index matrix, the comprehensive score of the undetermined flexible substrate material group is obtained and sorted to select primary flexible substrate materials in order.

[0108] Specifically, the present invention obtains deformation data of a pending flexible substrate material group within a curvature radius range of 3 to 8 mm, and calculates the curvature value by a three-point measurement method, wherein the measurement point spacing is set to 1 mm, and the sampling frequency is set to 100 Hz. A curvature displacement distribution cloud map is obtained according to the projection of the displacement field in the x-axis and y-axis directions, and the cloud map is used as sample data to train a convolutional neural network to obtain a deformation prediction model, wherein the number of channels of the model input layer is set to 2, including two channels in the x-direction and y-direction of the displacement field, and the hidden layer is 4 layers, and the feature map size is reduced from 64 to 8. The evolution law of the deformation variable is predicted under 1000 to 10000 cyclic bending conditions, and the deformation prediction value is judged by the displacement increment whether the deformation prediction value exceeds the preset deformation threshold value of 5%, and the data that does not exceed the threshold value is constructed as a deformation evolution curve of the pending flexible substrate material group under different curvature radii and bending times;

[0109] Obtain the resistance change rate data of the undetermined flexible substrate material group during the deformation process recorded by a four-probe resistance tester at a sampling frequency of 1000 Hz, and construct a fifth-order polynomial regressor based on the mapping relationship between the yield strength and the resistance change rate of the undetermined flexible substrate material group to calculate the quantitative correspondence between the resistance change rate and the deformation prediction quantity;

[0110] According to the resistance change rate data whose resistance change rate does not exceed the preset resistance change rate threshold and the deformation data in the deformation evolution curve and the quantitative correspondence between the two, a performance index matrix is ​​constructed, in which the row vector is the curvature radius and the number of bending times, and the column vector is the deformation variable and the resistance change amount. The curvature radius weight coefficient is set to 0.4 to reflect the importance of the bending resistance of the flexible substrate, the bending number weight coefficient is 0.3, the deformation variable weight coefficient is 0.2, and the resistance change weight coefficient is 0.1. The comprehensive score is calculated and sorted and screened, and the undetermined flexible substrate material group with a score not less than the preset threshold is used as the primary flexible substrate material.

[0111] Taking 5 different formulations of polyimide substrates as an example, the first formulation has a deformation of 3.2% and a resistance change of 7.5% after 10,000 cycles of bending at a radius of curvature of 8 mm; the second formulation has a deformation of 4.5% and a resistance change of 9.2%; the third formulation has a deformation of 4.8% and a resistance change of 9.8%; the fourth formulation has a deformation of 5.2% and a resistance change of 10.5%; the fifth formulation has a deformation of 5.5% and a resistance change of 11.2%. The comprehensive scores are calculated, and the first formulation has a highest score of 92 points, the second has 85 points, the third has 82 points, and the fourth and fifth formulations are screened out because they exceed the preset score threshold. It can be seen that the weight distribution highlights the impact of deformation control on the reliability of flexible substrates, and materials with smaller deformation variables get higher scores. The present invention constructs a performance index matrix including the radius of curvature, the number of bends, the deformation variable and the resistance change to comprehensively evaluate the comprehensive performance of the material, which not only improves the accuracy and objectivity of the evaluation, but also provides a scientific basis for subsequent material screening.

[0112] S4, constructing a feature mapping matrix according to the time-domain deformation data and the time-domain resistance change data of the primary flexible substrate material under multiple aging experiments, and constructing a conduction loss time-domain curve of the primary flexible substrate material during the aging process according to the feature mapping matrix;

[0113] In one embodiment, step S4 includes:

[0114] Acquire the time-domain deformation data of the primary flexible substrate material under multiple aging experiments to construct a deformation variable time-domain curve, and use a Wheatstone bridge to measure the time-domain resistance change data based on the deformation variable time-domain curve to construct a resistance change time-domain curve, and use the deformation variable time-domain curve as sample data to train a deep neural network to obtain the feature mapping matrix;

[0115] Based on the characteristic mapping matrix, the conduction loss value of the primary flexible substrate material is quantified by the Joule heat calculation formula to construct a conduction loss change curve, and the conduction loss change curve is decomposed on different time scales by using a three-layer wavelet transform to obtain three components of long-term trend, periodic fluctuation and random disturbance;

[0116] The prediction time span and confidence threshold are set, and the components are quantized by an autoregressor to obtain a time domain curve of the conduction loss of the primary flexible substrate material during the aging process.

[0117] Specifically, the present invention obtains the cumulative change of deformation variable with time in four time periods of 0 hours, 100 hours, 500 hours and 1000 hours under a constant temperature condition of 85 degrees Celsius applied to the primary flexible substrate material in an accelerated aging box, and collects the deformation data points in each time period, that is, the time-domain deformation data, in real time to obtain a standardized deformation variable time-domain curve; then, a Wheatstone bridge is used to measure the time-domain resistance change, the excitation voltage is set to 5 volts, the sampling frequency is 100 Hz, and the time-domain resistance change data is processed using a third-order polynomial fitter, and the fitting correlation coefficient reaches 0.98, to obtain a standardized evolution curve of the resistance change with time, that is, a resistance change time-domain curve; the resistance change time-domain curve and the deformation variable time-domain curve are used as sample data to train a deep neural network to obtain a feature mapping matrix, and the matrix uses a deep neural network with 4 nodes in the input layer, 8 nodes in the hidden layer, and 2 nodes in the output layer to train the relationship between the deformation variable and the resistance change;

[0118] Based on the feature mapping matrix, the conduction loss value of the primary flexible substrate material is quantified by the Joule heat calculation formula, and the part of the data exceeding the preset conduction loss threshold is eliminated to obtain the conduction loss change curve over time, that is, the conduction loss change curve, and the three-layer wavelet transform is used to decompose the conduction loss change curve on different time scales to obtain three components: long-term trend, periodic fluctuation and random disturbance; among them, the conduction loss calculation is based on the Joule heat formula, the current value is 100 mA, the conduction loss is 35 mW at 0 hour, increases to 38 mW at 100 hours, reaches 43 mW at 500 hours, and is 46 mW at 1000 hours, all of which do not exceed the 50 mW threshold;

[0119] The prediction time span is set to 2000 hours, the confidence threshold is 95%, and the three components are quantified by the autoregressor to obtain the change law of the conduction loss of the primary flexible substrate material during the long-term aging process, that is, the conduction loss time domain curve; among them, the trend component shows that the conduction loss presents a logarithmic growth characteristic, and the growth rate gradually decreases with time, and the autoregression prediction shows that at 2000 hours, the conduction loss will reach 48 milliwatts, with a confidence interval of 46 to 50 milliwatts. The periodic component reflects the impact of temperature fluctuations during the day and night on the conduction loss, with an amplitude of 1 milliwatt. The root mean square error between the decomposed and reconstructed conduction loss curve and the measured data is 0.8 milliwatts, which verifies the accuracy of the prediction model. The change law of conduction loss shows that the increase in conduction loss of the flexible substrate material within 2000 hours is within the expected range, indicating that it meets the long-term use requirements.

[0120] The present invention considers the time domain deformation data of the primary flexible substrate material in multiple aging experiments, which can accurately reflect the deformation of the material at different aging stages, and provides a solid foundation for subsequent analysis and prediction; by constructing a resistance change time domain curve, the change trend of the material resistance with the aging time can be intuitively displayed; by constructing a conduction loss change curve, the change trend of the material conduction loss with the aging time can be intuitively displayed, which is helpful to evaluate the aging degree and performance stability of the material; by decomposing the conduction loss change curve through a three-layer wavelet transform, a complex signal can be decomposed into a combination of different frequency components, so as to have a deeper understanding of the characteristics of the signal; combining a variety of technical means and tools, comprehensive consideration and optimization are carried out from data acquisition, processing, analysis to prediction, thereby improving the accuracy and reliability of the entire solution.

[0121] S5. constructing a multi-objective genetic optimization function according to the conduction loss time domain curve, and quantifying the material properties of the primary flexible substrate material based on the multi-objective genetic optimization function, and selecting a target flexible substrate material from the primary flexible substrate material according to the quantification result;

[0122] In one embodiment, step S5 includes:

[0123] The conduction loss time domain curve is subjected to piecewise linearization and standardization processing to obtain a piecewise standardized conduction loss function to construct the multi-objective genetic optimization function, and the parameters of the multi-objective genetic optimization function are optimized by genetic crossover and mutation operations to obtain a target conduction loss curve;

[0124] quantifying the root mean square error between the conduction loss time domain curve of the primary flexible substrate material and the target conduction loss curve by the least square method, and fitting the root mean square error by a radial basis kernel function support vector machine to quantify the material performance of the primary flexible substrate material and obtain a material performance score;

[0125] A screening matrix is ​​established according to the material performance score, and the loss growth rate and the loss baseline value are weighted by the screening matrix to calculate the weighted score of the primary flexible substrate material, and the primary flexible substrate material whose weighted score meets the weighted score requirement is used as the target flexible substrate material.

[0126] Specifically, the present invention performs piecewise linearization processing on the conduction loss time domain curve at 100-hour time intervals, and calculates the loss slope in each segment to obtain the loss growth rate, so as to divide the loss growth rate by a preset rate threshold of 50 milliwatts every 100 hours for standardization, and obtain a piecewise standardized conduction loss function with a time span of 1000 hours; and sets a multi-objective genetic optimization function according to the piecewise standardized conduction loss function, considering multiple optimization goals, such as reducing loss and improving stability, etc., wherein the optimizer population size is set to 100, the number of iterations is set to 1000 times, the convergence threshold is set to 0.001, the chromosome encoding includes two gene bits, namely the loss growth rate and the reference value, and offspring are generated through crossover operation, the crossover probability is set to 0.8, the mutation probability is set to 0.1, and the fitness function combines the growth rate weight of 0.6 and the reference value weight of 0.4 to obtain the target conduction loss curve, and the loss value of the sampling point at every 100 hours of the curve deviates from the measured value by less than 0.1 milliwatt;

[0127] For the target conduction loss curve, data is collected at sampling points at 100-hour intervals, and the root mean square error between the conduction loss time domain curve of each primary flexible substrate material and the target conduction loss curve is calculated using the least squares method. The root mean square error is fitted using a radial basis kernel support vector machine to quantify the material properties of the primary flexible substrate material and obtain a material performance score, which reflects the degree of compliance between the actual material performance and the target requirements. The support vector machine uses a radial basis kernel function, with the kernel parameter set to 0.1, and the optimal penalty factor of 100 is selected through cross-validation.

[0128] A screening matrix is ​​established based on the material performance scores. The loss growth rate weight of 0.6 and the loss baseline weight of 0.4 are set in the screening matrix. The loss growth rate weight of 0.6 in the material screening matrix reflects the long-term stability requirements, and the baseline value weight of 0.4 reflects the initial performance requirements. The weighted average score is calculated for each primary flexible substrate material. If the weighted score is greater than 90 points and the score fluctuation is less than 1 point for three consecutive optimization iterations, the material is determined to be the target material.

[0129] The present invention performs piecewise linearization and standardization on the conduction loss time domain curve, while maintaining the main features of the curve, it can also eliminate outliers and noise in the data, thereby improving the reliability and accuracy of the data; the multi-objective genetic optimization function can more comprehensively reflect the performance requirements of the material, and the parameters of the multi-objective genetic optimization function are optimized by genetic crossover and mutation operations, which can explore more solution spaces and improve the possibility of finding the optimal solution; based on the comprehensive evaluation of the screening matrix and the weighted score, it can ensure that the selected material reaches a high level in multiple performance indicators. The method is objective and repeatable, and can also be used for large-scale material screening and evaluation.

[0130] S6. quantifying the conduction loss data of the target flexible substrate material, and determining the conduction loss value of the target flexible device module according to the conduction loss data, so as to implement conduction loss analysis of the target flexible device module;

[0131] In one embodiment, the quantifying the conduction loss data of the target flexible substrate material includes:

[0132] Acquiring temperature field data of the target flexible substrate material to construct a temperature field distribution cloud map, and constructing a finite element calculation grid based on the temperature field distribution cloud map, so as to obtain the thermal conductivity and thermal diffusion coefficient of the target flexible substrate material through heat flux density calculation;

[0133] The thermal conductivity coefficient and the thermal diffusion coefficient are used as sample data to train a three-layer hidden layer neural network to obtain a thermal conductivity prediction model to output temperature distribution data, and according to the temperature distribution data, a power loss calculation formula based on Joule's heat law is used to quantify the conduction loss rate at each sampling point, so as to fit the temperature distribution data and the conduction loss rate through a third-order polynomial fitter to obtain a conduction loss distribution diagram;

[0134] The long short-term memory network is trained through the conduction loss distribution map to obtain a conduction loss optimization prediction model to output the power consumption change prediction trend, and when the power consumption change prediction trend meets the design index requirements of the target flexible device module, the power consumption change prediction trend is output as the conduction loss data of the target flexible substrate material.

[0135] Specifically, the temperature field data of the target flexible substrate material after a 5V DC power supply is applied is obtained, and the voltage drop and current value are measured at the same position using a four-probe voltage and current measuring instrument, and the on-state power consumption of each sampling point is calculated according to Joule's heat law to obtain a temperature field distribution cloud map and a current density distribution cloud map of the flexible substrate, that is, a temperature field distribution cloud map; a finite element calculation grid is constructed according to the temperature field distribution cloud map using a 0.1mm grid size, the boundary temperature is set to 25 degrees Celsius, and the thermal convection coefficient is 15 watts per square meter Kelvin, and the thermal conductivity and thermal diffusion coefficient of the target flexible substrate material are obtained by heat flux density calculation;

[0136] A three-layer hidden layer neural network is used to train 1000 sets of temperature field data and their corresponding thermal conductivity and thermal diffusion coefficient. The number of hidden layer nodes is 64, 32, and 16 respectively. A thermal conductivity prediction model is obtained to output temperature distribution data. Based on the data, the conduction loss rate at each sampling point is calculated using the power loss calculation formula based on Joule's heat law, that is, the ratio of power consumption per unit area to input power. The relationship between temperature and power consumption changes is fitted by a third-order polynomial fitter. The fitting correlation coefficient is set to 0.98 to obtain a conduction loss distribution diagram. In the neural network training, 1000 sets of temperature field data are input, including temperature distribution under different working conditions, and temperature field features are extracted through three hidden layers.

[0137] The long short-term memory network is trained according to the conduction loss distribution map to obtain the conduction loss optimization prediction model. The maximum loss rate threshold is set to 0.2, and the temperature uniformity threshold is set to 5 degrees Celsius. The conduction loss optimization prediction model is used to predict the power consumption change trend in the next 1000 hours. The network input dimension is 16, the hidden layer dimension is 32, the output dimension is 1, and the time step is 100 hours. When the power consumption change prediction trend meets the design index requirements of the target flexible device module, it is output as the conduction loss data of the target flexible substrate material, that is, as the conduction analysis result output of the target flexible substrate material.

[0138] The present invention comprehensively considers the mechanical and electrical properties of the flexible substrate material and the performance changes during the aging process to screen the target material corresponding to the target flexible device module, thereby improving the accuracy of the conduction loss assessment; it adopts advanced calculation methods and tools, such as finite element analysis, neural network prediction, long short-term memory network, etc., to improve the calculation efficiency and prediction accuracy.

[0139] In addition, after obtaining the conduction loss data of the target flexible substrate material corresponding to the target flexible device module, and calculating the conduction loss of each component and connector contained in the target flexible device module, the conduction loss of the target flexible substrate material, electronic components and connectors are finally added together to obtain the total conduction loss of the target flexible device module, so as to realize the conduction loss analysis of the target flexible device module; wherein, for each electronic component and connector, its conduction loss in the circuit contained in the target flexible device module can be calculated according to its resistance and current value.

[0140] In the embodiment of the present application, based on the problem of how to improve the accuracy of the existing conduction loss analysis method, a conduction loss analysis method for a flexible device module is designed, which realizes obtaining the performance parameters of the target flexible device module, and selecting a to-be-determined flexible substrate material group from a plurality of flexible substrate materials according to the performance parameters; constructing a mapping relationship between the mechanical characteristics and the electrical characteristics of the to-be-determined flexible substrate material group according to the force changes and resistance changes of the to-be-determined flexible substrate material group under a plurality of load conditions; constructing a deformation evolution curve based on the deformation data of the to-be-determined flexible substrate material group under a plurality of curvature radii, and generating a performance index matrix in combination with the mapping relationship, so as to select a primary flexible substrate material from the to-be-determined flexible substrate material group; and selecting a primary flexible substrate material from the to-be-determined flexible substrate material group according to the time domain deformation data and the time domain deformation data of the primary flexible substrate material under multiple aging experiments. A feature mapping matrix is ​​constructed based on domain resistance change data, and a time domain curve of the conduction loss of the primary flexible substrate material during the aging process is constructed according to the feature mapping matrix; a multi-objective genetic optimization function is constructed according to the conduction loss time domain curve, and the material properties of the primary flexible substrate material are quantified based on the multi-objective genetic optimization function, and a target flexible substrate material is selected from the primary flexible substrate materials according to the quantification result; the conduction loss data of the target flexible substrate material is quantified, and the conduction loss value of the target flexible device module is determined according to the conduction loss data, so as to realize a technical solution for conducting loss analysis of the target flexible device module; the mechanical and electrical properties of the flexible substrate material and the performance changes during the aging process are comprehensively considered, so as to realize accurate analysis of the conduction loss of the flexible device module.

[0141] It should be noted that although the steps in the above flowchart are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders.

[0142] In another embodiment, if Figure 2 As shown, the second aspect of the present invention provides a conduction loss analysis system for a flexible device module, comprising:

[0143] A range determination module 10 is used to obtain performance parameters of a target flexible device module and select a to-be-determined flexible substrate material group from a plurality of flexible substrate materials according to the performance parameters;

[0144] A relationship building module 20, for building a mapping relationship between the mechanical characteristics and the electrical characteristics of the to-be-determined flexible substrate material group according to the force changes and resistance changes of the to-be-determined flexible substrate material group under various load conditions;

[0145] A primary screening module 30 is used to construct a deformation evolution curve based on the deformation data of the undetermined flexible substrate material group under several curvature radii, and generate a performance index matrix in combination with the mapping relationship to select a primary flexible substrate material from the undetermined flexible substrate material group;

[0146] A curve construction module 40 is used to construct a characteristic mapping matrix according to the time-domain deformation data and the time-domain resistance change data of the primary flexible substrate material under multiple aging experiments, and to construct a conduction loss time-domain curve of the primary flexible substrate material during the aging process according to the characteristic mapping matrix;

[0147] a target selection module 50, configured to construct a multi-objective genetic optimization function according to the conduction loss time domain curve, quantify the material properties of the primary flexible substrate material based on the multi-objective genetic optimization function, and select a target flexible substrate material from the primary flexible substrate material according to the quantification result;

[0148] The loss analysis module 60 is used to quantify the conduction loss data of the target flexible substrate material and determine the conduction loss value of the target flexible device module according to the conduction loss data to implement conduction loss analysis of the target flexible device module.

[0149] It should be noted that each module in the above-mentioned conduction loss analysis system for flexible device modules can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. For the specific definition of a conduction loss analysis system for flexible device modules, please refer to the definition of a conduction loss analysis method for flexible device modules above. The two have the same functions and effects, which will not be repeated here.

[0150] In summary, the present invention relates to the field of information technology, and discloses a conduction loss analysis method and system for flexible device modules. The screening range of flexible substrate materials is determined by the performance parameters of the target flexible device module, and the deformation evolution curve and the correlation between mechanical and electrical characteristics are established according to the stress and resistance changes of the flexible device materials to be determined within the range under various loads. The two are combined to form a performance index matrix for preliminary screening for screening; using the aging experimental data of the materials after the preliminary screening, a characteristic matrix reflecting the change of material performance over time is constructed, and the time domain curve of the conduction loss is further obtained; through a multi-objective genetic optimization function, the performance of the primary materials is quantitatively scored to select the optimal target material; based on the conduction loss data of the target material, and the mechanical and electrical characteristics of the flexible substrate material and the performance changes during the aging process are comprehensively considered, the conduction loss of the flexible device module is accurately analyzed.

[0151] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0152] The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.

Claims

1. A conduction loss analysis method for a flexible device module, characterized in that: include: Acquiring performance parameters of a target flexible device module, and selecting a to-be-determined flexible substrate material group from a plurality of flexible substrate materials according to the performance parameters; According to the force changes and resistance changes of the to-be-determined flexible substrate material group under various load conditions, a mapping relationship between the mechanical characteristics and the electrical characteristics of the to-be-determined flexible substrate material group is constructed; Constructing a deformation evolution curve based on deformation data of the undetermined flexible substrate material group under several curvature radii, and generating a performance index matrix in combination with the mapping relationship, so as to select a primary flexible substrate material from the undetermined flexible substrate material group; Constructing a characteristic mapping matrix according to the time-domain deformation data and the time-domain resistance change data of the primary flexible substrate material under multiple aging experiments, and constructing a conduction loss time-domain curve of the primary flexible substrate material during the aging process according to the characteristic mapping matrix; Constructing a multi-objective genetic optimization function according to the conduction loss time domain curve, and quantifying the material properties of the primary flexible substrate material based on the multi-objective genetic optimization function, and selecting a target flexible substrate material from the primary flexible substrate material according to the quantification result; Quantifying the conduction loss data of the target flexible substrate material, and determining the conduction loss value of the target flexible device module according to the conduction loss data, so as to implement conduction loss analysis of the target flexible device module; The method of constructing a multi-objective genetic optimization function according to the conduction loss time domain curve, quantifying the material properties of the primary flexible substrate material based on the multi-objective genetic optimization function, and selecting a target flexible substrate material from the primary flexible substrate material according to the quantification result includes: The conduction loss time domain curve is subjected to piecewise linearization and standardization processing to obtain a piecewise standardized conduction loss function to construct the multi-objective genetic optimization function, and the parameters of the multi-objective genetic optimization function are optimized by genetic crossover and mutation operations to obtain a target conduction loss curve; quantifying the root mean square error between the conduction loss time domain curve of the primary flexible substrate material and the target conduction loss curve by the least square method, and fitting the root mean square error by a radial basis kernel function support vector machine to quantify the material performance of the primary flexible substrate material and obtain a material performance score; A screening matrix is ​​established according to the material performance score, and the loss growth rate and the loss baseline value are weighted by the screening matrix to calculate the weighted score of the primary flexible substrate material, and the primary flexible substrate material whose weighted score meets the weighted score requirement is used as the target flexible substrate material.

2. A conduction loss analysis method for flexible device modules according to claim 1, characterized in that: The selecting a to-be-determined flexible substrate material group from a plurality of flexible substrate materials according to the performance parameters comprises: Determine the electrical conductivity threshold, thermal conductivity threshold, stress threshold, strain threshold, interlayer bonding strength threshold and fatigue resistance threshold of the flexible substrate material according to the performance parameters; Acquiring the electrical conductivity and thermal conductivity of the plurality of flexible substrate materials, and selecting the flexible substrate materials whose electrical conductivity exceeds the electrical conductivity threshold and whose corresponding thermal conductivity exceeds the thermal conductivity threshold as a pre-selected material group; Calculating stress data and strain data of the preselected material group under a preset external force by a finite element analysis method, and taking the preselected material group whose stress data exceeds the stress threshold and whose corresponding strain data exceeds the strain threshold as a candidate material group; Obtaining interlayer bonding strength data of the candidate material group and fatigue resistance values ​​obtained by performing a cyclic bending test on the candidate material group, and selecting a candidate material group whose interlayer bonding strength data exceeds the interlayer bonding strength threshold and whose corresponding fatigue resistance values ​​exceed the fatigue resistance threshold as a candidate material group; The adaptability score of the candidate material group is quantified under the coupling of thermal stress and mechanical stress, and the candidate material group that meets the adaptability requirement is used as the to-be-determined flexible substrate material group.

3. A conduction loss analysis method for flexible device modules according to claim 2, characterized in that: The step of quantifying the adaptability score of the candidate material group under the coupling of thermal stress and mechanical stress comprises: Obtaining the thermal expansion coefficient, strain allowable range and Young's modulus of the candidate material group to quantify the stress state of the candidate material group under the coupling of thermal stress and mechanical stress, and obtaining mechanical stress and thermal stress; The mechanical stress and the thermal stress are combined by finite element analysis to obtain a coupled stress field, and the safety margin, stress concentration factor and fatigue life of the alternative material group are evaluated based on the coupled stress field to quantify the adaptability score of the alternative material group by a weighted average method.

4. The conduction loss analysis method for flexible device modules according to claim 1, characterized in that: The step of constructing a mapping relationship between the mechanical characteristics and the electrical characteristics of the flexible substrate material group to be determined according to the force changes and resistance changes of the flexible substrate material group to be determined under various load conditions includes: The mechanical parameters of the to-be-determined flexible substrate material group are collected to construct a tetrahedral unit space grid, and the grid cells in the tetrahedral unit space grid are refined by boundary point constraints to obtain a tetrahedral grid calculation graph; Obtaining a displacement cloud map and a strain field distribution map after applying multiple load conditions to the tetrahedral grid calculation map, and extracting the displacement and strain distribution data of the undetermined flexible substrate material group after deformation based on the displacement cloud map and the strain field distribution map, to obtain a stress-strain curve of the undetermined flexible substrate material group; Substituting the displacement and the strain distribution data into the Pearson resistance-strain relationship formula to calculate the conductivity gradient of the undetermined flexible substrate material group, and spatially interpolating the conductivity gradient through a deep neural network to obtain conductivity distribution data to construct a local load compensation matrix; Based on the local load compensation matrix, a support vector regressor is used to calculate the resistance change of the undetermined flexible substrate material group under various load conditions to obtain a resistance change curve, and a mapping relationship between the mechanical characteristics and the electrical characteristics of the undetermined flexible substrate material group is constructed according to the resistance change curve and the stress-strain curve.

5. A conduction loss analysis method for flexible device modules according to claim 4, characterized in that: The step of constructing a mapping relationship between the mechanical characteristics and the electrical characteristics of the to-be-determined flexible substrate material group according to the resistance change curve and the stress-strain curve includes: Fitting the stress-strain curve by the least square method, and determining the turning point position of the elastic deformation and plastic deformation of the flexible substrate material group to be determined by the first-order derivative change rate of the fitting curve, so as to obtain standardized mechanical characteristic parameters; Based on the resistance change curve, the resistance change rate is calculated by sliding the time window, and the time domain characteristic value of the resistance change rate is collected to obtain the standardized electrical characteristic parameters of the to-be-determined flexible substrate material group; The standardized mechanical characteristic parameters and the standardized electrical characteristic parameters are used as sample data to train a neural network model, and the associated weights of stress strain and resistance change are added between the hidden layer nodes of the neural network model, and the weight gradient is calculated by back propagation to obtain a trained neural network model; The parameters of the trained neural network model are optimized through a radial basis function support vector machine regressor, and the optimal parameter combination is selected by cross-validation. When the prediction error of the neural network model after parameter adjustment reaches the error accuracy, the mapping relationship between the mechanical characteristics and electrical characteristics of the undetermined flexible substrate material group is obtained.

6. The conduction loss analysis method for flexible device modules according to claim 1, characterized in that: The step of constructing a deformation evolution curve based on the deformation data of the undetermined flexible substrate material group under several curvature radii, and combining the deformation evolution curve with the mapping relationship to generate a performance index matrix to select a primary flexible substrate material from the undetermined flexible substrate material group includes: Obtaining deformation data of the undetermined flexible substrate material group under several curvature radii, obtaining a curvature displacement distribution cloud map to train a convolutional neural network as sample data, obtaining a deformation prediction model to output a predicted deformation amount of the undetermined flexible substrate material group under several cyclic bending conditions, and constructing the deformation evolution curve based on the predicted deformation amount; Obtaining resistance change rate data of the to-be-determined flexible substrate material group at several curvature radii, and constructing a fifth-order polynomial regressor based on the mapping relationship to calculate the quantitative correspondence between the resistance change rate data and the data in the deformation evolution curve; According to the quantitative correspondence, a performance index matrix is ​​constructed, wherein the row vectors are the radius of curvature and the number of bending times, and the column vectors are the deformation amount and the resistance change amount. By performing weighted calculation on the performance index matrix, the comprehensive score of the undetermined flexible substrate material group is obtained and sorted to select primary flexible substrate materials in order.

7. The conduction loss analysis method for flexible device modules according to claim 1, characterized in that: The method of constructing a characteristic mapping matrix according to the time-domain deformation data and the time-domain resistance change data of the primary flexible substrate material under multiple aging experiments, and constructing a conduction loss time-domain curve of the primary flexible substrate material during the aging process according to the characteristic mapping matrix, includes: Acquire the time-domain deformation data of the primary flexible substrate material under multiple aging experiments to construct a deformation variable time-domain curve, and use a Wheatstone bridge to measure the time-domain resistance change data based on the deformation variable time-domain curve to construct a resistance change time-domain curve, and use the deformation variable time-domain curve as sample data to train a deep neural network to obtain the feature mapping matrix; Based on the characteristic mapping matrix, the conduction loss value of the primary flexible substrate material is quantified by the Joule heat calculation formula to construct a conduction loss change curve, and the conduction loss change curve is decomposed on different time scales by using a three-layer wavelet transform to obtain three components of long-term trend, periodic fluctuation and random disturbance; The prediction time span and confidence threshold are set, and the components are quantized by an autoregressor to obtain a time domain curve of the conduction loss of the primary flexible substrate material during the aging process.

8. The conduction loss analysis method for flexible device modules according to claim 1, characterized in that: The quantifying the conduction loss data of the target flexible substrate material includes: Acquiring temperature field data of the target flexible substrate material to construct a temperature field distribution cloud map, and constructing a finite element calculation grid based on the temperature field distribution cloud map, so as to obtain the thermal conductivity and thermal diffusion coefficient of the target flexible substrate material through heat flux density calculation; The thermal conductivity coefficient and the thermal diffusion coefficient are used as sample data to train a three-layer hidden layer neural network to obtain a thermal conductivity prediction model to output temperature distribution data, and according to the temperature distribution data, a power loss calculation formula based on Joule's heat law is used to quantify the conduction loss rate at each sampling point, so as to fit the temperature distribution data and the conduction loss rate through a third-order polynomial fitter to obtain a conduction loss distribution diagram; The long short-term memory network is trained through the conduction loss distribution map to obtain a conduction loss optimization prediction model to output the power consumption change prediction trend, and when the power consumption change prediction trend meets the design index requirements of the target flexible device module, the power consumption change prediction trend is output as the conduction loss data of the target flexible substrate material.

9. A conduction loss analysis system for flexible device modules, characterized in that: include: A range determination module, used for obtaining performance parameters of a target flexible device module, and selecting a to-be-determined flexible substrate material group from a plurality of flexible substrate materials according to the performance parameters; A relationship building module, used to build a mapping relationship between the mechanical characteristics and electrical characteristics of the to-be-determined flexible substrate material group according to the force changes and resistance changes of the to-be-determined flexible substrate material group under various load conditions; A primary screening module, for constructing a deformation evolution curve based on deformation data of the undetermined flexible substrate material group under several curvature radii, and generating a performance index matrix in combination with the mapping relationship, so as to select a primary flexible substrate material from the undetermined flexible substrate material group; A curve construction module, used to construct a feature mapping matrix according to the time-domain deformation data and the time-domain resistance change data of the primary flexible substrate material under multiple aging experiments, and to construct a conduction loss time-domain curve of the primary flexible substrate material during the aging process according to the feature mapping matrix; a target selection module, configured to construct a multi-objective genetic optimization function according to the conduction loss time domain curve, quantify the material properties of the primary flexible substrate material based on the multi-objective genetic optimization function, and select a target flexible substrate material from the primary flexible substrate material according to the quantification result; A loss analysis module, used to quantify the conduction loss data of the target flexible substrate material, and determine the conduction loss value of the target flexible device module according to the conduction loss data, so as to realize the conduction loss analysis of the target flexible device module; The method of constructing a multi-objective genetic optimization function according to the conduction loss time domain curve, quantifying the material properties of the primary flexible substrate material based on the multi-objective genetic optimization function, and selecting a target flexible substrate material from the primary flexible substrate material according to the quantification result includes: The conduction loss time domain curve is subjected to piecewise linearization and standardization processing to obtain a piecewise standardized conduction loss function to construct the multi-objective genetic optimization function, and the parameters of the multi-objective genetic optimization function are optimized by genetic crossover and mutation operations to obtain a target conduction loss curve; quantifying the root mean square error between the conduction loss time domain curve of the primary flexible substrate material and the target conduction loss curve by the least square method, and fitting the root mean square error by a radial basis kernel function support vector machine to quantify the material performance of the primary flexible substrate material and obtain a material performance score; A screening matrix is ​​established according to the material performance score, and the loss growth rate and the loss baseline value are weighted by the screening matrix to calculate the weighted score of the primary flexible substrate material, and the primary flexible substrate material whose weighted score meets the weighted score requirement is used as the target flexible substrate material.

Images