Preparation optimization method and system of heat-conducting solid crystal glue

Abstract

The invention provides a preparation optimization method and system of a heat-conducting solid crystal adhesive, and relates to the technical field of solid crystal adhesives, and the preparation optimization method comprises the following steps: collecting specification parameters and energy consumption parameters of a chip which is subjected to solid crystal by adopting a to-be-prepared solid crystal adhesive; the solid crystal glue component is optimized, the optimal solid crystal glue component is obtained, and optimization is carried out with the optimization purposes of improving the shape stability of the solid crystal glue and reducing the dispensing difficulty of the solid crystal glue; carrying out optimization selection on the filling amount interval of the silver nanowires in the die bonding glue to obtain the filling amount interval of the silver nanowires; in the silver nanowire filling amount interval and the decentralized control parameter range, the filling amount and the decentralized control parameters of the silver nanowires are optimized, and the optimal filling amount and the optimal decentralized control parameters are obtained; and preparing the solid crystal glue by adopting the optimal solid crystal glue components, the filling amount and the decentralized control parameters. The technical problems that a traditional preparation parameter optimization method usually needs a large number of tests and adjustment, the preparation period is long, and the production efficiency is low are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of die-bonding adhesives, and in particular to a preparation optimization method and system of a thermally conductive die-bonding adhesive. Background Art

[0002] Thermally conductive adhesive is a widely used material, mainly used to fix and conduct heat to electronic chips, protect circuit structures, and improve heat dissipation efficiency. Its quality and performance directly affect the reliability and stability of electronic equipment. However, there are still some technical problems in the traditional preparation method of thermally conductive adhesive. On the one hand, the traditional preparation method of adhesive often cannot accurately control the preparation parameters of adhesive, resulting in fluctuations and instability in the performance of adhesive, affecting the working effect of electronic equipment; on the other hand, the traditional preparation parameter optimization method usually requires a lot of experiments and adjustments to obtain the ideal preparation parameters, resulting in a long preparation cycle and low production efficiency. Summary of the invention

[0003] The present application provides a method for optimizing the preparation of a thermally conductive bonding adhesive, aiming to solve the technical problems that traditional bonding adhesive preparation methods cannot accurately control preparation parameters, resulting in fluctuations in the performance of the bonding adhesive, and traditional preparation parameter optimization methods usually require a large number of experiments and adjustments, resulting in long preparation cycles and low production efficiency.

[0004] In view of the above problems, the present application provides a method and system for optimizing the preparation of a thermally conductive die-bonding adhesive.

[0005] The first aspect disclosed in the present application provides a preparation optimization method for a thermally conductive bonding adhesive, the method comprising: collecting specification parameters and energy consumption parameters of a chip to be bonded with the bonding adhesive to be prepared; optimizing the composition of the bonding adhesive according to the specification parameters to obtain an optimal bonding adhesive composition, wherein the optimization is performed according to the specification parameters with the purpose of improving the shape stability of the bonding adhesive and reducing the difficulty of dispensing the bonding adhesive; optimizing and selecting the silver nanowire filling amount interval in the bonding adhesive based on the optimal bonding adhesive composition and energy consumption parameters to obtain a silver nanowire filling amount interval; optimizing the silver nanowire filling amount and dispersion control parameters according to the energy consumption parameters and the optimal bonding adhesive composition within the silver nanowire filling amount interval and the dispersion control parameter range to obtain an optimal filling amount and an optimal dispersion control parameter; and preparing the bonding adhesive using the optimal bonding adhesive composition, the optimal filling amount and the optimal dispersion control parameter.

[0006] In a second aspect disclosed in the present application, a system for optimizing the preparation of a thermally conductive bonding adhesive is provided, the system being used for the above-mentioned method for optimizing the preparation of a thermally conductive bonding adhesive, the system comprising: a parameter acquisition module, the parameter acquisition module being used to collect specification parameters and energy consumption parameters of a chip that is bonded with the bonding adhesive to be prepared; a component optimization module, the component optimization module being used to optimize the composition of the bonding adhesive according to the specification parameters to obtain an optimal bonding adhesive composition, wherein the optimization is performed according to the specification parameters with the purpose of improving the shape stability of the bonding adhesive and reducing the difficulty of dispensing the bonding adhesive; an optimization selection module, the The optimization selection module is used to optimize the silver nanowire filling amount interval in the solid crystal glue based on the optimal solid crystal glue composition and energy consumption parameters to obtain the silver nanowire filling amount interval; the parameter optimization module is used to optimize the silver nanowire filling amount and dispersion control parameters within the silver nanowire filling amount interval and dispersion control parameter range according to the energy consumption parameters and the optimal solid crystal glue composition to obtain the optimal filling amount and the optimal dispersion control parameters; the solid crystal glue preparation module is used to prepare the solid crystal glue using the optimal solid crystal glue composition, the optimal filling amount and the optimal dispersion control parameters.

[0007] According to a third aspect disclosed in the present application, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, any step of the first aspect disclosed in the present application is implemented.

[0008] The fourth aspect disclosed in the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the first aspect disclosed in the present application.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] By optimizing the composition of the solid crystal glue according to the specification parameters, the solid crystal glue has better shape stability and reduced dispensing difficulty under different specifications and application scenarios. This customized optimization of the composition of the solid crystal glue can meet the needs of different chips and improve the quality of solid crystal and production efficiency; by optimizing the silver nanowire filling amount and dispersion control parameters, the thermal conductivity of the solid crystal glue is optimized. The optimized filling amount and dispersion control parameters make the thermal conductive solid crystal glue have better thermal conductivity, while reducing costs and improving production efficiency; by adopting a systematic optimization method, the scientific optimization of the solid crystal glue composition, silver nanowire filling amount, and dispersion control parameters is achieved, the controllability and stability of the preparation process are improved, the trial and error cost is reduced, and the production efficiency is improved. In summary, the preparation optimization method of the thermal conductive solid crystal glue realizes the multi-faceted optimization of the preparation process of the thermal conductive solid crystal glue, including the optimization of the composition of the solid crystal glue, the silver nanowire filling amount, and the dispersion control parameters, so that the thermal conductive solid crystal glue has better performance and economy.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of a process flow of a method for optimizing the preparation of a thermally conductive die-bonding adhesive provided in an embodiment of the present application;

[0013] Figure 2 A schematic diagram of the structure of a system for optimizing the preparation of a thermally conductive die-bonding adhesive provided in an embodiment of the present application;

[0014] Figure 3 This is a diagram of the internal structure of a computer device provided in an embodiment of the present application.

[0015] Explanation of reference numerals: parameter acquisition module 10 , component optimization module 20 , optimization selection module 30 , parameter optimization module 40 , die-bonding adhesive preparation module 50 . DETAILED DESCRIPTION

[0016] The embodiment of the present application provides a method for optimizing the preparation of a thermally conductive bonding adhesive, thereby solving the technical problems that the traditional bonding adhesive preparation method cannot accurately control the preparation parameters, resulting in fluctuations in the performance of the bonding adhesive, and the traditional preparation parameter optimization method usually requires a large number of experiments and adjustments, resulting in a long preparation cycle and low production efficiency.

[0017] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0018] like Figure 1 As shown, an embodiment of the present application provides a method for optimizing the preparation of a thermally conductive die-bonding adhesive, the method comprising:

[0019] Collecting specification parameters and energy consumption parameters of the chip to be bonded with the bonding glue to be prepared;

[0020] Collect chip samples for which the die-bonding adhesive is to be prepared, including chips of different specifications and sizes. These samples cover various situations that may be encountered in application scenarios. Measure the specification parameters of the collected chip samples, including the size of the chip, including length, width, thickness, and the layout of the heat dissipation components on the chip. These parameters are used for subsequent optimization of the die-bonding adhesive composition and performance evaluation.

[0021] After the chip is die-sealed, the energy consumption parameters of the chip are recorded. For example, the energy consumption of the chip under different workloads and environmental conditions is obtained through actual testing or simulation. The energy consumption parameters include power consumption, temperature change, operating frequency, etc.

[0022] By obtaining the specification parameters and energy consumption parameters of the chip to be bonded with the bonding adhesive to be prepared, basic data is provided for the subsequent optimization of the bonding adhesive composition and performance improvement.

[0023] According to the specification parameters, the composition of the crystal bonding adhesive is optimized to obtain the optimal crystal bonding adhesive composition, wherein, according to the specification parameters, the optimization is performed with the purpose of improving the shape stability of the crystal bonding adhesive and reducing the difficulty of dispensing the crystal bonding adhesive;

[0024] Analyze the components of the solid crystal adhesive, including resin, filler, diluent, etc., to obtain the properties, functions and impact of each component on the solid crystal adhesive. According to the specification parameters, in order to improve the shape stability of the solid crystal adhesive, adjust the ratio of resin and filler in the solid crystal adhesive, and select components with good shape retention ability to improve the stability of the solid crystal adhesive; at the same time, in order to reduce the difficulty of dispensing the solid crystal adhesive, it can be achieved by adjusting the type and content of the diluent, as well as the selection of resin. The optimization method is used to optimize the solid crystal adhesive composition within the component ratio range of the solid crystal adhesive, and finally generate the optimal solid crystal adhesive composition. The optimization steps are only briefly mentioned here, and the specific implementation process is detailed in the subsequent steps.

[0025] Furthermore, according to the specification parameters, the composition of the die-bonding adhesive is optimized to obtain the optimal die-bonding adhesive composition, including:

[0026] Construct a composition optimization function to optimize the composition of the die-bonding adhesive, as follows:

[0027]

[0028] Among them, SCA is the component fitness, w1 is the stability weight, w2 is the difficulty weight, X r is the shape stability grade of the die-bonding adhesive composition, X y is the preset shape stability level, N r N is the difficulty level of dispensing under the composition of die-bonding glue, y It is the preset dispensing difficulty level;

[0029] Obtaining the component ratio range of multiple components in the die-bonding adhesive;

[0030] Randomly generate a die-bonding adhesive component within the component ratio range, analyze and obtain the shape stability level and the dispensing difficulty level in combination with the specification parameters, and calculate and obtain the component fitness according to the component optimization function;

[0031] Continue to optimize the composition of the bonding adhesive to obtain the optimal bonding adhesive composition with the greatest adaptability.

[0032] Construct a composition optimization function to optimize the composition of the die-bonding adhesive, as follows:

[0033]

[0034] This composition optimization function is used to evaluate the quality of the bonding adhesive composition. It mainly considers two aspects, namely, shape stability and dispensing difficulty under the bonding adhesive composition. Among them, w1 is the stability weight and w2 is the difficulty weight. These weights are set according to actual needs and are used to adjust the importance of different aspects in the optimization process. For example, if shape stability is more important for the application, w1 is set higher; X y The preset shape stability level is the expected shape stability level, which is determined based on the needs of a specific application or product; N y The preset dispensing difficulty level is the expected dispensing difficulty level, which is determined according to the application requirements and manufacturing conditions; X r N is the shape stability grade of the die-bonding adhesive composition. This grade is derived from the evaluation of the shape stability of the die-bonding adhesive under the selected combination of components. r It is the dispensing difficulty level of the die-bonding adhesive under the selected component combination. This level is derived from the dispensing difficulty evaluation of the die-bonding adhesive under the selected component combination.

[0035] Based on the known components of the die-bonding adhesive and their ranges, the ratio range of various components, such as resin, filler and diluent, is determined. This range can be determined based on the properties of the components, application requirements and limitations in the preparation process.

[0036] Within a certain component ratio range, multiple combinations of bonding adhesive components are randomly generated. For each randomly generated bonding adhesive component combination, the influence on shape stability and dispensing difficulty is analyzed in combination with the specification parameters. This can be evaluated through experiments, simulations, or based on existing data. Using the constructed component optimization function, according to the influence of each component combination on shape stability and dispensing difficulty, the shape stability level and dispensing difficulty level are substituted into the component optimization function for calculation to obtain the component fitness of each combination.

[0037] Repeat the above process until all the combinations of bonding adhesive components are traversed within the component ratio range, and the component fitness of each combination is obtained. From all the generated bonding adhesive component combinations, the combination with the largest component fitness is selected as the optimal bonding adhesive component. This combination corresponds to the optimal shape stability and dispensing difficulty, thus providing a basis for subsequent bonding adhesive preparation work.

[0038] Furthermore, in combination with the specification parameters, the shape stability level and dispensing difficulty level are analyzed and obtained, including:

[0039] According to the composition of the die-bonding adhesive, a shape stability grade of the die-bonding adhesive sample is obtained by testing;

[0040] According to the dispensing and bonding data records of the die-bonding adhesive, a sample die-bonding adhesive component set and a sample specification parameter set are obtained, and according to the dispensing success rate of chips with different specification parameters, a sample dispensing difficulty level set is obtained;

[0041] The sample die-bonding adhesive component set, the sample specification parameter set and the sample dispensing difficulty level set are used to construct a dispensing difficulty analyzer, and the dispensing difficulty analysis is performed on the specification parameters and the die-bonding adhesive components to obtain the dispensing difficulty level.

[0042] According to the randomly generated bonding adhesive composition, bonding adhesive samples are prepared, and the shape stability of the bonding adhesive samples is tested using appropriate testing methods and equipment. For example, a microscope is used to observe whether the shape is deformed, or a measuring instrument is used to measure the size and geometric characteristics of the shape. According to the test results, the shape stability of the bonding adhesive samples is divided into different levels, such as excellent, good, general, and poor. This can be determined based on indicators such as the degree of shape deformation and dimensional deviation of the sample, and ultimately the shape stability level of the bonding adhesive sample is obtained.

[0043] Collect dispensing and bonding data records of different combinations of bonding adhesives, including bonding adhesive composition, specification parameters, dispensing success rate, etc. Extract the component combination set and sample specification parameter set of the bonding adhesive from the dispensing and bonding data records of the bonding adhesive.

[0044] According to the dispensing success rate of chips with different specifications and parameters, the samples are divided into different dispensing difficulty levels. For example, samples with high success rate can be divided into low dispensing difficulty levels, and samples with low success rate can be divided into high dispensing difficulty levels. Record the sample die-bonding glue component set, sample specification parameter set, and sample dispensing difficulty level set for subsequent analysis and optimization.

[0045] Integrate the set of sample solid crystal adhesive components, the set of sample specification parameters, and the set of sample dispensing difficulty levels into a data set, wherein the sample solid crystal adhesive components and sample specification parameters are encoded as numerical features of the machine learning model, and the dispensing difficulty level is encoded as a label of the machine learning model. Divide the data set into a training set and a test set. Select a suitable machine learning model, such as a neural network, a support vector machine, etc. Use the training set to train the selected machine learning model. During the training process, the model learns the relationship between the solid crystal adhesive components, specification parameters, and the dispensing difficulty level. Use the test set to evaluate the performance of the model. The evaluation indicators include accuracy, precision, etc. Tune the model according to the evaluation results, and finally obtain a dispensing difficulty analyzer that meets the preset accuracy, which is used to perform dispensing difficulty analysis on specification parameters and solid crystal adhesive components.

[0046] The trained dispensing difficulty analyzer is used to take the specification parameters and the composition of the die-bonding adhesive as input to predict the dispensing difficulty level. The dispensing difficulty level is obtained based on the prediction results of the model.

[0047] Based on the optimal bonding adhesive composition and energy consumption parameters, optimizing and selecting the silver nanowire filling amount interval in the bonding adhesive to obtain the silver nanowire filling amount interval;

[0048] Furthermore, based on the optimal bonding adhesive composition and energy consumption parameters, the silver nanowire filling amount interval in the bonding adhesive is optimized and selected to obtain the silver nanowire filling amount interval, including:

[0049] Obtaining a sample solid crystal adhesive component set and a sample first silver nanowire filling amount set, and constructing a first filling amount mapping relationship;

[0050] Mapping the optimal bonding adhesive composition within the first filling amount mapping relationship, obtaining a plurality of sample first silver nanowire filling amounts corresponding to a plurality of sample bonding adhesive compositions that are closest to each other, and constructing a first filling amount interval;

[0051] According to the chip bonding data record, a sample energy consumption parameter set and a sample second silver nanowire filling amount set are obtained, and a second filling amount mapping relationship is constructed;

[0052] Mapping the energy consumption parameter in the second filling amount mapping relationship, obtaining a plurality of second silver nanowire filling amounts of samples corresponding to the most similar energy consumption parameters of the samples, and constructing a second filling amount interval;

[0053] The silver nanowire filling amount interval is constructed by combining the first filling amount interval and the second filling amount interval.

[0054] According to the dispensing and bonding data records of the bonding adhesive, a set of sample bonding adhesive components is obtained, and a set of sample first silver nanowire filling amounts is extracted from the data records. These data include the silver nanowire filling amounts under different combinations of bonding adhesive components. For each sample bonding adhesive component, there is a corresponding sample first silver nanowire filling amount. These data are organized into a mapping table, in which the bonding adhesive component is used as a key and the first silver nanowire filling amount is used as a value, and a first filling amount mapping relationship is established.

[0055] The determined optimal bonding adhesive composition is used as input, matched in the first filling amount mapping relationship, and multiple sample bonding adhesive compositions closest to the optimal bonding adhesive composition are found, and the first silver nanowire filling amounts of multiple samples corresponding to them are obtained. The obtained first silver nanowire filling amounts of multiple samples are statistically analyzed to determine their range and distribution, and determine the boundaries of the filling amount interval, so as to construct the first filling amount interval.

[0056] According to the chip bonding data record, a sample energy consumption parameter set is obtained, and a sample second silver nanowire filling amount set is extracted from the data record. These data include the silver nanowire filling amount under different energy consumption parameters. For each sample energy consumption parameter, there is a corresponding sample second silver nanowire filling amount. These data are organized into a mapping table, in which the sample energy consumption parameter is used as the key and the second silver nanowire filling amount is used as the value, and a second filling amount mapping relationship is established.

[0057] Using the determined energy consumption parameter as input, matching in the second filling amount mapping relationship, finding multiple sample energy consumption parameters closest to the energy consumption parameter, and obtaining the second silver nanowire filling amounts of multiple samples corresponding to them. Performing statistical analysis on the obtained second silver nanowire filling amounts of multiple samples, determining their range and distribution, determining the boundary of the filling amount interval, and thereby constructing the second filling amount interval.

[0058] The first filling amount interval and the second filling amount interval are combined, that is, the first filling amount interval and the second filling amount interval are intersected to determine the final silver nanowire filling amount interval.

[0059] Within the silver nanowire filling amount interval and the dispersion control parameter range, according to the energy consumption parameter and the optimal solid crystal adhesive composition, the silver nanowire filling amount and the dispersion control parameter are optimized to obtain the optimal filling amount and the optimal dispersion control parameter;

[0060] Furthermore, within the silver nanowire filling amount interval and the dispersion control parameter range, the silver nanowire filling amount and the dispersion control parameter are optimized according to the energy consumption parameter and the optimal solid crystal adhesive composition to obtain the optimal filling amount and the optimal dispersion control parameter, including:

[0061] obtaining a dispersion control parameter range for performing dispersion control on the silver nanowires;

[0062] According to the specification parameters and energy consumption parameters, multiple historical filling amounts and historical decentralized control parameters of the die-bonding glue for the same type of chips during the historical period of time are obtained by indexing, and multiple target solutions are constructed as the first generation of target solution clusters;

[0063] Randomly generate multiple initial solutions within the silver nanowire filling amount interval and the dispersion control parameter range to obtain a first generation of initial solution clusters, and randomly construct a mapping relationship with multiple target solutions in the target solution cluster, wherein each initial solution includes a filling amount and a dispersion control parameter;

[0064] Taking the target solution cluster as the optimization direction, adjusting and updating multiple initial solutions in the initial solution cluster according to the mapping relationship, obtaining multiple updated solutions, and constructing an updated solution cluster;

[0065] Performing random cross-update on the filling amounts or decentralized control parameters in the updated solutions and the target solutions in the updated solution cluster and the target solution cluster to obtain multiple cultivated solutions and construct a cultivated solution cluster;

[0066] According to the filling amounts and dispersion control parameters in the multiple cultivation solutions, the thermal conductivity parameters and filling costs of the die-bonding adhesive are analyzed, and in combination with the energy consumption parameters, multiple cultivation filling fitnesses are calculated and obtained;

[0067] Analyze and calculate to obtain target filling fitness of multiple target solutions, and combine the multiple cultivation filling fitness to discriminate and update the target solution cluster;

[0068] Based on the updated target solution cluster, a new initial solution cluster is randomly generated, and iterative optimization is performed until convergence. The filling amount and decentralized control parameters with the maximum filling fitness are output as the optimal filling amount and optimal decentralized control parameters.

[0069] Determine the various parameters used to control the dispersion of silver nanowires, including ultrasonic frequency, ultrasonic power, stirring speed, stirring time, etc. For each dispersion control parameter, determine its range of variation, which can be determined through experimental data and literature research. Finally, determine the dispersion control parameter range of silver nanowires to provide a reference for subsequent optimization.

[0070] Based on the determined specification parameters and energy consumption parameters as index conditions, the chips with the same specification parameters and energy consumption parameters within the historical time range are retrieved as the same type of chips, and their bonding glue data records are obtained. Multiple historical filling amounts and historical decentralized control parameters are extracted from the retrieved historical bonding glue data records, and these data correspond to the chip specification parameters and energy consumption parameters. The extracted multiple historical filling amounts and historical decentralized control parameters are combined into multiple target solutions as the first generation of target solution clusters.

[0071] In the silver nanowire filling amount interval and the dispersion control parameter range, the filling amount and the dispersion control parameter are randomly selected to generate multiple initial solutions, which are combined into the first generation of initial solution clusters. The mapping relationship between multiple initial solutions in the initial solution cluster and multiple target solutions in the target solution cluster is randomly constructed, that is, a target solution is randomly assigned to each initial solution in the initial solution cluster so that it can be optimized corresponding to the target solution during the optimization process.

[0072] The target solution cluster is taken as the optimization direction, and each initial solution in the initial solution cluster is adjusted and updated according to the established mapping relationship. Specifically, for each initial solution, its corresponding target solution is determined according to the mapping relationship, and the target solution is used as the reference direction for adjustment and update. After adjusting and updating each initial solution, multiple updated solutions are obtained, and the multiple updated solutions obtained are combined into an updated solution cluster.

[0073] Multiple updated solutions and multiple target solutions are extracted from the updated solution cluster and the target solution cluster. For each selected solution, the filling amount or decentralized control parameter is randomly selected for cross-updating. The cross-update can adopt various cross-updating methods, such as single-point cross-over, multi-point cross-over or uniform cross-over. After the filling amount or decentralized control parameter of each solution is randomly cross-updated, multiple new breeding solutions are obtained, and the multiple breeding solutions obtained are combined into a breeding solution cluster.

[0074] Using the filling amounts and dispersion control parameters in multiple cultivation solutions, calculate the thermal conductivity parameters of the solid crystal glue, including thermal conductivity, thermal resistance, etc.; calculate the filling cost of the solid crystal glue, including material cost, preparation process cost, etc. According to the thermal conductivity parameters and filling cost, combined with the energy consumption parameters, a filling function is used to calculate the filling fitness for each cultivation solution to obtain multiple cultivation filling fitnesses. The specific thermal conductivity parameters and filling cost acquisition methods, and the cultivation filling fitness calculation methods are described in detail in the subsequent steps and are only briefly mentioned here.

[0075] For each target solution in the target solution cluster, the target filling fitness is calculated by the same method as the aforementioned cultivation filling fitness calculation to obtain multiple target filling fitnesses of multiple target solutions. The calculated multiple cultivation filling fitnesses are compared with the multiple target filling fitnesses, and the target solution cluster is discriminated and updated according to the comparison result, for example, according to the size of the filling fitness, the target solution is selected to be retained, updated or eliminated, so as to retain the target solution that meets the requirements and update or eliminate the unqualified target solution.

[0076] Based on the updated target solution cluster, new initial solution clusters are randomly generated, which contain multiple different combinations of filling amount and decentralized control parameters. The newly generated initial solution clusters are iteratively optimized. In each iteration, the filling fitness of each solution is calculated, and the parameters of the solution are updated according to the optimization algorithm. After each round of iteration, it is checked whether the convergence condition is met. The convergence condition can be that the change of filling fitness is less than the change threshold, or the preset number of iterations is reached. When the convergence condition is met, the iteration is stopped, and the solution with the largest filling fitness is selected from all optimized solutions as the optimal solution, and the filling amount and decentralized control parameters corresponding to the optimal solution are output as the optimal filling amount and optimal decentralized control parameters.

[0077] Furthermore, performing random cross-update on the filling amounts or decentralized control parameters in the updated solutions and the target solutions in the updated solution cluster and the target solution cluster comprises:

[0078] According to the mapping relationship, the deviation coefficients of the filling amount and the decentralized control parameter in the updated solution and the filling amount and the decentralized control parameter in the target solution that are mapped to each other are calculated to obtain a plurality of filling amount deviation coefficients and a plurality of decentralized deviation coefficients;

[0079] According to the multiple filling amount deviation coefficients and the multiple dispersion deviation coefficients, multiple crossover probability distributions are obtained by allocation calculation, wherein each crossover probability distribution includes a filling amount probability interval and a dispersion probability interval;

[0080] Randomly generate random numbers from 0 to 1, and judge the probability intervals in which the multiple crossover probability distributions fall respectively. According to the multiple probability intervals, cross-update the filling amounts or decentralized control parameters in the multiple update solutions and the multiple target solutions to obtain multiple training solutions.

[0081] For each pair of updated solutions and target solutions that are mapped to each other, calculate the deviation coefficient between their filling amounts, for example, calculate the difference between the filling amount of the updated solution and the filling amount of the target solution, and then divide the difference by the filling amount of the target solution to obtain the filling amount deviation coefficient; similarly, for each pair of updated solutions and target solutions, calculate the deviation coefficient between their decentralized control parameters, for example, calculate the difference between the decentralized control parameters of the updated solution and the decentralized control parameters of the target solution, and then divide the difference by the decentralized control parameters of the target solution to obtain the decentralized deviation coefficient. Traverse all combinations of updated solutions and target solutions to obtain multiple filling amount deviation coefficients and multiple decentralized deviation coefficients.

[0082] For each filling amount deviation coefficient and dispersion deviation coefficient, calculate its absolute value, which ensures that all deviations are taken into account, not just positive or negative deviations. The absolute value of the filling amount deviation coefficient and the absolute value of the dispersion deviation coefficient are used as weights of the probability distribution. The larger the absolute value of the deviation coefficient, the higher the corresponding probability weight. Using the assigned probability weights, calculate the filling amount probability interval and the dispersion probability interval. For example, normalize the probability weights to a sum of 1, and then determine the probability distribution of the filling amount and the dispersion control parameters based on the normalized probability weights. For example, the deviation coefficients of the filling amount and the dispersion control parameters in a pair of updated solutions and target solutions are 0.2 and 0.3. First normalize them to 0.4 and 0.6, then the crossover probability distribution includes a filling amount probability interval of 0 to 0.4 and a dispersion probability interval of 0.4 to 1.

[0083] The fill amount probability interval and the dispersion probability interval are combined to generate multiple crossover probability distributions, each of which includes probability distributions of the fill amount and the dispersion control parameter.

[0084] A random number between 0 and 1 is randomly generated, and according to the probability intervals of the filling amount or the probability intervals of the dispersion control parameters of multiple crossover probability distributions, it is determined which probability interval the random number falls into, and the corresponding crossover probability distribution is selected according to the probability interval the random number falls into.

[0085] According to the selected crossover probability distribution, the filling amount or decentralized control parameters in multiple update solutions and multiple target solutions are cross-updated. The crossover update can adopt some common methods, such as single-point crossover, multi-point crossover or uniform crossover, etc. After the crossover update, multiple new cultivation solutions are obtained. Through the judgment of random numbers and probability intervals, as well as the operation of crossover update, multiple diverse cultivation solutions can be obtained, which helps to better explore the parameter space during the optimization process and improve the efficiency and accuracy of optimization.

[0086] Furthermore, according to the filling amount and dispersion control parameters in the multiple cultivation solutions, the thermal conductivity parameters and filling cost of the die-bonding adhesive are analyzed, and combined with the energy consumption parameters, multiple cultivation filling fitnesses are calculated, including:

[0087] According to the filling preparation data record of the die-bonding glue, a sample die-bonding glue component set, a sample filling amount set and a sample dispersion control parameter set are obtained, and a sample thermal conductivity performance parameter set and a sample filling cost set are obtained;

[0088] The sample die-bonding adhesive composition set, the sample filling amount set and the sample dispersion control parameter set are used as input, and the sample thermal conductivity performance parameter set and the sample filling cost set are used as output to construct a filling analyzer, and the filling amounts and dispersion control parameters in the multiple cultivation solutions are analyzed in combination with the optimal die-bonding adhesive composition to obtain multiple thermal conductivity performance parameters and multiple filling costs;

[0089] Construct the filling function as follows:

[0090]

[0091] Among them, SF is the filling fitness, w3 is the thermal conductivity weight, w4 is the cost weight, D is the thermal conductivity performance parameter of the die-bonding adhesive under the optimal die-bonding adhesive composition, filling amount and dispersion control parameters, P is the energy consumption parameter, M r is the filling amount and the filling cost of the decentralized control parameter, M y Fill the cost for the preset;

[0092] According to the filling function, according to the energy consumption parameter, multiple thermal conductivity parameters and multiple filling costs, multiple cultivation filling fitnesses are calculated and obtained.

[0093] Obtain the filling preparation data record of the die-bonding adhesive, and extract the sample die-bonding adhesive composition set, sample filling amount set, and sample dispersion control parameter set from the filling preparation data record. These samples cover different die-bonding adhesive compositions, filling amounts, and dispersion control parameters. Extract the thermal conductivity performance parameters of the die-bonding adhesive, such as thermal conductivity, from the filling preparation data record to form a sample thermal conductivity performance parameter set; extract filling cost information, such as material cost, preparation process cost, etc., to form a sample filling cost set.

[0094] The sample die-bonding adhesive composition set, sample filling amount set and sample dispersion control parameter set are used as the input features of the model, and the sample thermal conductivity parameter set and sample filling cost set are used as the output labels of the model to obtain the constructed data set. Select a suitable model structure, such as a neural network, support vector machine, etc.

[0095] The constructed data set is divided into a training set and a test set. The selected model is trained using the training set. The model parameters are adjusted by minimizing the loss function. The performance of the trained model is evaluated using the test set. Evaluation indicators such as mean square error and mean absolute error are used to evaluate the predictive ability of the model. The model is tuned according to the evaluation results until the preset conditions are met, such as the preset accuracy. A trained filling analyzer is obtained to predict the thermal conductivity parameters and filling cost of new samples.

[0096] Construct the filling function as follows:

[0097]

[0098] This filling function is used to evaluate the fitness of the filling solution, taking into account two key factors, namely thermal conductivity and filling cost. Among them, the filling fitness SF is the evaluation index of the filling solution, which indicates the quality of the filling solution under specific conditions; w3 is the thermal conductivity weight, which is a weight parameter used to adjust the importance of thermal conductivity in the filling fitness. A larger weight indicates that the thermal conductivity has a greater impact on the fitness; w4 is the cost weight, which is a weight parameter used to adjust the importance of the filling cost in the filling fitness. A larger weight indicates that the filling cost has a greater impact on the fitness; M y It is the preset filling cost and is used to compare the filling costs.

[0099] The first term of the filling function considers the relationship between thermal conductivity and energy consumption, that is, the higher the thermal conductivity and the lower the energy consumption, the higher the filling fitness. The second term considers the relationship between the filling cost and the preset filling cost, that is, the lower the filling cost, the higher the filling fitness. By combining these two terms, the filling function can evaluate the comprehensive performance of the filling solution in terms of thermal conductivity and cost.

[0100] For each cultivation solution, its thermal conductivity parameters, filling cost, and energy consumption parameters are imported into the filling function for calculation. The calculated value is the filling fitness of each cultivation solution. After calculation, multiple cultivation filling fitnesses are obtained. The filling fitness reflects the comprehensive performance of the cultivation solution in terms of thermal conductivity and cost, and is used to evaluate its quality.

[0101] The optimal bonding adhesive composition, optimal filling amount and optimal dispersion control parameters are used to prepare the bonding adhesive.

[0102] The optimal bonding adhesive composition, optimal filling amount and optimal dispersion control parameters obtained by previous optimization are used to prepare the bonding adhesive to obtain a bonding adhesive with excellent performance.

[0103] In summary, the preparation optimization method of a thermal conductive die-bonding adhesive provided in the embodiment of the present application has the following technical effects:

[0104] 1. By optimizing the composition of the die-bonding adhesive according to the specification parameters, the die-bonding adhesive has better shape stability and reduced dispensing difficulty in different specifications and application scenarios. This customized optimization of the die-bonding adhesive composition can meet the needs of different chips and improve the quality of die bonding and production efficiency;

[0105] 2. The thermal conductivity of the die-bonding adhesive is optimized by optimizing the silver nanowire filling amount and dispersion control parameters. The optimized filling amount and dispersion control parameters make the thermal conductive die-bonding adhesive have better thermal conductivity, while reducing costs and improving production efficiency;

[0106] 3. By adopting a systematic optimization method, the scientific optimization of the solid crystal glue composition, silver nanowire filling amount, and dispersion control parameters was achieved, which improved the controllability and stability of the preparation process, reduced the trial and error cost, and improved production efficiency.

[0107] In summary, the preparation optimization method of the thermal conductive bonding adhesive realizes the multi-faceted optimization of the preparation process of the thermal conductive bonding adhesive, including the optimization of the bonding adhesive composition, the silver nanowire filling amount, and the dispersion control parameters, so that the thermal conductive bonding adhesive has better performance and economy.

[0108] Based on the same inventive concept as the preparation and optimization method of a thermal conductive die-bonding adhesive in the aforementioned embodiment, Figure 2 As shown, the present application provides a preparation optimization system for thermal conductive die-bonding adhesive, the system comprising:

[0109] A parameter acquisition module 10, the parameter acquisition module 10 is used to collect specification parameters and energy consumption parameters of a chip that is bonded with the die-bonding adhesive to be prepared;

[0110] A component optimization module 20, the component optimization module 20 is used to optimize the components of the crystal bonding adhesive according to the specification parameters to obtain the optimal crystal bonding adhesive components, wherein the optimization is performed according to the specification parameters to improve the shape stability of the crystal bonding adhesive and reduce the difficulty of dispensing the crystal bonding adhesive;

[0111] An optimization selection module 30, the optimization selection module 30 is used to optimize the silver nanowire filling amount interval in the crystal bonding adhesive based on the optimal crystal bonding adhesive composition and energy consumption parameters to obtain the silver nanowire filling amount interval;

[0112] A parameter optimization module 40, wherein the parameter optimization module 40 is used to optimize the silver nanowire filling amount and dispersion control parameters according to the energy consumption parameters and the optimal solid crystal adhesive composition within the silver nanowire filling amount interval and the dispersion control parameter range to obtain an optimal filling amount and an optimal dispersion control parameter;

[0113] The bonding adhesive preparation module 50 is used to prepare the bonding adhesive by using the optimal bonding adhesive composition, optimal filling amount and optimal dispersion control parameters.

[0114] Furthermore, the system further includes an optimal bonding adhesive composition acquisition module to perform the following operation steps:

[0115] Construct a composition optimization function to optimize the composition of the die-bonding adhesive, as follows:

[0116]

[0117] Among them, SCA is the component fitness, w1 is the stability weight, w2 is the difficulty weight, X r is the shape stability grade of the die-bonding adhesive composition, X y is the preset shape stability level, N r N is the difficulty level of dispensing under the composition of die-bonding glue, y It is the preset dispensing difficulty level;

[0118] Obtaining the component ratio range of multiple components in the die-bonding adhesive;

[0119] Randomly generate a die-bonding adhesive component within the component ratio range, analyze and obtain the shape stability level and the dispensing difficulty level in combination with the specification parameters, and calculate and obtain the component fitness according to the component optimization function;

[0120] Continue to optimize the composition of the bonding adhesive to obtain the optimal bonding adhesive composition with the greatest adaptability.

[0121] Furthermore, the system further includes a dispensing difficulty level acquisition module to perform the following operation steps:

[0122] According to the composition of the die-bonding adhesive, a shape stability grade of the die-bonding adhesive sample is obtained by testing;

[0123] According to the dispensing and bonding data records of the die-bonding adhesive, a sample die-bonding adhesive component set and a sample specification parameter set are obtained, and according to the dispensing success rate of chips with different specification parameters, a sample dispensing difficulty level set is obtained;

[0124] The sample die-bonding adhesive component set, the sample specification parameter set and the sample dispensing difficulty level set are used to construct a dispensing difficulty analyzer, and the dispensing difficulty analysis is performed on the specification parameters and the die-bonding adhesive components to obtain the dispensing difficulty level.

[0125] Furthermore, the system further includes a silver nanowire filling amount interval acquisition module to perform the following operation steps:

[0126] Obtaining a sample solid crystal adhesive component set and a sample first silver nanowire filling amount set, and constructing a first filling amount mapping relationship;

[0127] Mapping the optimal bonding adhesive composition within the first filling amount mapping relationship, obtaining a plurality of sample first silver nanowire filling amounts corresponding to a plurality of sample bonding adhesive compositions that are closest to each other, and constructing a first filling amount interval;

[0128] According to the chip bonding data record, a sample energy consumption parameter set and a sample second silver nanowire filling amount set are obtained, and a second filling amount mapping relationship is constructed;

[0129] Mapping the energy consumption parameter in the second filling amount mapping relationship, obtaining a plurality of second silver nanowire filling amounts of samples corresponding to the most similar energy consumption parameters of the samples, and constructing a second filling amount interval;

[0130] The silver nanowire filling amount interval is constructed by combining the first filling amount interval and the second filling amount interval.

[0131] Furthermore, the system further includes an optimal decentralized control parameter acquisition module to perform the following operation steps:

[0132] obtaining a dispersion control parameter range for performing dispersion control on the silver nanowires;

[0133] According to the specification parameters and energy consumption parameters, multiple historical filling amounts and historical decentralized control parameters of the die-bonding glue for the same type of chips during the historical period of time are obtained by indexing, and multiple target solutions are constructed as the first generation of target solution clusters;

[0134] Randomly generate multiple initial solutions within the silver nanowire filling amount interval and the dispersion control parameter range to obtain a first generation of initial solution clusters, and randomly construct a mapping relationship with multiple target solutions in the target solution cluster, wherein each initial solution includes a filling amount and a dispersion control parameter;

[0135] Taking the target solution cluster as the optimization direction, adjusting and updating multiple initial solutions in the initial solution cluster according to the mapping relationship, obtaining multiple updated solutions, and constructing an updated solution cluster;

[0136] Performing random cross-update on the filling amounts or decentralized control parameters in the updated solutions and the target solutions in the updated solution cluster and the target solution cluster to obtain multiple cultivated solutions and construct a cultivated solution cluster;

[0137] According to the filling amounts and dispersion control parameters in the multiple cultivation solutions, the thermal conductivity parameters and filling costs of the die-bonding adhesive are analyzed, and in combination with the energy consumption parameters, multiple cultivation filling fitnesses are calculated and obtained;

[0138] Analyze and calculate to obtain target filling fitness of multiple target solutions, and combine the multiple cultivation filling fitness to discriminate and update the target solution cluster;

[0139] Based on the updated target solution cluster, a new initial solution cluster is randomly generated, and iterative optimization is performed until convergence. The filling amount and decentralized control parameters with the maximum filling fitness are output as the optimal filling amount and optimal decentralized control parameters.

[0140] Furthermore, the system further comprises a plurality of training solution acquisition modules to perform the following operation steps:

[0141] According to the mapping relationship, the deviation coefficients of the filling amount and the decentralized control parameter in the updated solution and the filling amount and the decentralized control parameter in the target solution that are mapped to each other are calculated to obtain a plurality of filling amount deviation coefficients and a plurality of decentralized deviation coefficients;

[0142] According to the multiple filling amount deviation coefficients and the multiple dispersion deviation coefficients, multiple crossover probability distributions are obtained by allocation calculation, wherein each crossover probability distribution includes a filling amount probability interval and a dispersion probability interval;

[0143] Randomly generate random numbers from 0 to 1, and judge the probability intervals in which the multiple crossover probability distributions fall respectively. According to the multiple probability intervals, cross-update the filling amounts or decentralized control parameters in the multiple update solutions and the multiple target solutions to obtain multiple training solutions.

[0144] Furthermore, the system also includes a plurality of cultivation and filling fitness acquisition modules to perform the following operation steps:

[0145] According to the filling preparation data record of the die-bonding glue, a sample die-bonding glue component set, a sample filling amount set and a sample dispersion control parameter set are obtained, and a sample thermal conductivity performance parameter set and a sample filling cost set are obtained;

[0146] The sample die-bonding adhesive composition set, the sample filling amount set and the sample dispersion control parameter set are used as input, and the sample thermal conductivity performance parameter set and the sample filling cost set are used as output to construct a filling analyzer, and the filling amounts and dispersion control parameters in the multiple cultivation solutions are analyzed in combination with the optimal die-bonding adhesive composition to obtain multiple thermal conductivity performance parameters and multiple filling costs;

[0147] Construct the filling function as follows:

[0148]

[0149] Among them, SF is the filling fitness, w3 is the thermal conductivity weight, w4 is the cost weight, D is the thermal conductivity performance parameter of the die-bonding adhesive under the optimal die-bonding adhesive composition, filling amount and dispersion control parameters, P is the energy consumption parameter, M ris the filling amount and the filling cost of the decentralized control parameter, M y Fill the cost for the preset;

[0150] According to the filling function, according to the energy consumption parameter, multiple thermal conductivity parameters and multiple filling costs, multiple cultivation filling fitnesses are calculated and obtained.

[0151] Through the above-mentioned detailed description of the preparation optimization method of a thermally conductive bonding adhesive in this specification, those skilled in the art can clearly know the preparation optimization system of a thermally conductive bonding adhesive in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0152] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the processor of the computer device is used to provide computing and control capabilities; the memory of the computer device includes a non-volatile storage medium and an internal memory, wherein the non-volatile storage medium stores an operating system, a computer program and a database, and the internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium; the network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a method for optimizing the preparation of a thermal conductive die-bonding adhesive.

[0153] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0154] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are 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.

[0155] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for optimizing the preparation of a thermally conductive die-bonding adhesive, characterized in that: The method comprises: Collecting specification parameters and energy consumption parameters of the chip to be bonded with the bonding glue to be prepared; According to the specification parameters, the composition of the crystal bonding adhesive is optimized to obtain the optimal crystal bonding adhesive composition, wherein, according to the specification parameters, the optimization is performed with the purpose of improving the shape stability of the crystal bonding adhesive and reducing the difficulty of dispensing the crystal bonding adhesive; Based on the optimal bonding adhesive composition and energy consumption parameters, optimizing and selecting the silver nanowire filling amount interval in the bonding adhesive to obtain the silver nanowire filling amount interval; Within the silver nanowire filling amount interval and the dispersion control parameter range, according to the energy consumption parameter and the optimal solid crystal adhesive composition, the silver nanowire filling amount and the dispersion control parameter are optimized to obtain the optimal filling amount and the optimal dispersion control parameter; The optimal bonding adhesive composition, optimal filling amount and optimal dispersion control parameters are used to prepare the bonding adhesive.

2. The method according to claim 1, characterized in that According to the specification parameters, the composition of the die-bonding adhesive is optimized to obtain the optimal die-bonding adhesive composition, including: Construct a composition optimization function to optimize the composition of the die-bonding adhesive, as follows: Among them, SCA is the component fitness, w1 is the stability weight, w2 is the difficulty weight, X r is the shape stability grade under the die-bonding adhesive composition, X y is the preset shape stability level, N r N is the difficulty level of dispensing under the composition of die-bonding glue, y It is the preset dispensing difficulty level; Obtaining the component ratio range of multiple components in the die-bonding adhesive; Randomly generate a die-bonding adhesive component within the component ratio range, analyze and obtain the shape stability level and the dispensing difficulty level in combination with the specification parameters, and calculate and obtain the component fitness according to the component optimization function; Continue to optimize the composition of the bonding adhesive to obtain the optimal bonding adhesive composition with the greatest component adaptability.

3. The method according to claim 2, characterized in that Combined with the above specifications, the shape stability level and dispensing difficulty level are analyzed and obtained, including: According to the composition of the die-bonding adhesive, a shape stability grade of the die-bonding adhesive sample is obtained by testing; According to the dispensing and bonding data records of the die-bonding adhesive, a sample die-bonding adhesive component set and a sample specification parameter set are obtained, and according to the dispensing success rate of chips with different specification parameters, a sample dispensing difficulty level set is obtained; The sample die-bonding adhesive component set, the sample specification parameter set and the sample dispensing difficulty level set are used to construct a dispensing difficulty analyzer, and the dispensing difficulty analysis is performed on the specification parameters and the die-bonding adhesive components to obtain the dispensing difficulty level.

4. The method according to claim 1, characterized in that: Based on the optimal bonding adhesive composition and energy consumption parameters, the silver nanowire filling amount interval in the bonding adhesive is optimized and selected to obtain the silver nanowire filling amount interval, including: Obtaining a sample solid crystal adhesive component set and a sample first silver nanowire filling amount set, and constructing a first filling amount mapping relationship; Mapping the optimal bonding adhesive composition within the first filling amount mapping relationship, obtaining a plurality of sample first silver nanowire filling amounts corresponding to a plurality of sample bonding adhesive compositions that are closest to each other, and constructing a first filling amount interval; According to the chip bonding data record, a sample energy consumption parameter set and a sample second silver nanowire filling amount set are obtained, and a second filling amount mapping relationship is constructed; Mapping the energy consumption parameter in the second filling amount mapping relationship, obtaining a plurality of second silver nanowire filling amounts of samples corresponding to the most similar energy consumption parameters of the samples, and constructing a second filling amount interval; The silver nanowire filling amount interval is constructed by combining the first filling amount interval and the second filling amount interval.

5. The method according to claim 1, characterized in that Within the silver nanowire filling amount interval and the dispersion control parameter range, the silver nanowire filling amount and the dispersion control parameter are optimized according to the energy consumption parameter and the optimal solid crystal adhesive composition to obtain the optimal filling amount and the optimal dispersion control parameter, including: obtaining a dispersion control parameter range for performing dispersion control on the silver nanowires; According to the specification parameters and energy consumption parameters, multiple historical filling amounts and historical decentralized control parameters of the die-bonding glue for the same type of chips during the historical period of time are obtained by indexing, and multiple target solutions are constructed as the first generation of target solution clusters; Randomly generate multiple initial solutions within the silver nanowire filling amount interval and the dispersion control parameter range to obtain a first generation of initial solution clusters, and randomly construct a mapping relationship with multiple target solutions in the target solution cluster, wherein each initial solution includes a filling amount and a dispersion control parameter; Taking the target solution cluster as the optimization direction, adjusting and updating multiple initial solutions in the initial solution cluster according to the mapping relationship, obtaining multiple updated solutions, and constructing an updated solution cluster; Performing random cross-update on the filling amounts or decentralized control parameters in the updated solutions and the target solutions in the updated solution cluster and the target solution cluster to obtain multiple cultivated solutions and construct a cultivated solution cluster; According to the filling amounts and dispersion control parameters in the multiple cultivation solutions, the thermal conductivity parameters and filling costs of the die-bonding adhesive are analyzed, and in combination with the energy consumption parameters, multiple cultivation filling fitnesses are calculated and obtained; Analyze and calculate to obtain target filling fitness of multiple target solutions, and combine the multiple cultivation filling fitness to discriminate and update the target solution cluster; Based on the updated target solution cluster, a new initial solution cluster is randomly generated, and iterative optimization is performed until convergence. The filling amount and decentralized control parameters with the maximum filling fitness are output as the optimal filling amount and optimal decentralized control parameters.

6. The method according to claim 5, characterized in that The filling amounts or decentralized control parameters in the updated solutions and the target solutions in the updated solution cluster and the target solution cluster are randomly cross-updated, including: According to the mapping relationship, the deviation coefficients of the filling amount and the decentralized control parameter in the updated solution and the filling amount and the decentralized control parameter in the target solution that are mapped to each other are calculated to obtain a plurality of filling amount deviation coefficients and a plurality of decentralized deviation coefficients; According to the multiple filling amount deviation coefficients and the multiple dispersion deviation coefficients, multiple crossover probability distributions are obtained by allocation calculation, wherein each crossover probability distribution includes a filling amount probability interval and a dispersion probability interval; Randomly generate random numbers from 0 to 1, and judge the probability intervals in which the multiple crossover probability distributions fall respectively. According to the multiple probability intervals, cross-update the filling amounts or decentralized control parameters in the multiple update solutions and the multiple target solutions to obtain multiple training solutions.

7. The method according to claim 5, characterized in that According to the filling amounts and dispersion control parameters in the multiple cultivation solutions, the thermal conductivity parameters and filling costs of the die-bonding adhesive are analyzed, and in combination with the energy consumption parameters, multiple cultivation filling fitnesses are calculated, including: According to the filling preparation data record of the die-bonding glue, a sample die-bonding glue component set, a sample filling amount set and a sample dispersion control parameter set are obtained, and a sample thermal conductivity performance parameter set and a sample filling cost set are obtained; The sample die-bonding adhesive composition set, the sample filling amount set and the sample dispersion control parameter set are used as input, and the sample thermal conductivity performance parameter set and the sample filling cost set are used as output to construct a filling analyzer, and the filling amounts and dispersion control parameters in the multiple cultivation solutions are analyzed in combination with the optimal die-bonding adhesive composition to obtain multiple thermal conductivity performance parameters and multiple filling costs; Construct the filling function as follows: Among them, SF is the filling fitness, w3 is the thermal conductivity weight, w4 is the cost weight, D is the thermal conductivity performance parameter of the die-bonding adhesive under the optimal die-bonding adhesive composition, filling amount and dispersion control parameters, P is the energy consumption parameter, M r is the filling amount and the filling cost of the decentralized control parameter, M y Fill the cost for the preset; According to the filling function, according to the energy consumption parameter, multiple thermal conductivity parameters and multiple filling costs, multiple cultivation filling fitnesses are calculated and obtained.

8. A preparation optimization system for thermal conductive die-bonding adhesive, characterized in that: A method for optimizing the preparation of a thermally conductive die-bonding adhesive according to any one of claims 1 to 7, the system comprising: A parameter acquisition module, the parameter acquisition module is used to collect specification parameters and energy consumption parameters of the chip that is bonded with the die-bonding adhesive to be prepared; A component optimization module, the component optimization module is used to optimize the components of the crystal bonding adhesive according to the specification parameters to obtain the optimal crystal bonding adhesive components, wherein the optimization is performed according to the specification parameters to improve the shape stability of the crystal bonding adhesive and reduce the difficulty of dispensing the crystal bonding adhesive; An optimization selection module, the optimization selection module is used to optimize the silver nanowire filling amount interval in the crystal bonding adhesive based on the optimal crystal bonding adhesive composition and energy consumption parameters to obtain the silver nanowire filling amount interval; A parameter optimization module, wherein the parameter optimization module is used to optimize the silver nanowire filling amount and dispersion control parameters according to the energy consumption parameters and the optimal solid crystal adhesive composition within the silver nanowire filling amount interval and the dispersion control parameter range to obtain the optimal filling amount and the optimal dispersion control parameters; The bonding adhesive preparation module is used to prepare the bonding adhesive by using the optimal bonding adhesive composition, optimal filling amount and optimal dispersion control parameters.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for optimizing the preparation of a thermally conductive die-bonding adhesive according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for optimizing the preparation of a thermally conductive die-bonding adhesive according to any one of claims 1 to 7 are implemented.

Images