Big data driven intelligent gold bonding wire system performance prediction method and system

Abstract

The present invention relates to the field of materials science and engineering, and discloses a big data driven intelligent gold bonding wire system performance prediction method, comprising: constructing a gold wire physical model of the gold bonding wire, defining a data driven network architecture of the gold bonding wire, and determining a data driven model of the gold bonding wire; defining a GANs network of the intelligent gold bonding wire system, analyzing the gold wire feature tuple of the gold bonding wire, training the gold wire physical model and the data driven model, and obtaining a training gold wire physical model and a training data driven model; analyzing the physical model performance and driving model performance of the training gold wire physical model and the training data driven model, coupling the training gold wire physical model and the training data driven model, and obtaining a hybrid gold wire analysis model; capturing the performance factor dependency of the gold bonding wire, analyzing the potential performance coefficient of the gold bonding wire using the hybrid gold wire analysis model, and constructing a performance analysis report of the gold bonding wire. The present invention can improve the performance analysis effect of the gold bonding wire.

Description

Technical Field

[0001] The present invention relates to a big data driven intelligent gold bonding wire system performance prediction method and system, belonging to the field of material science and engineering. Background Art

[0002] Intelligent gold bonding wire system performance prediction refers to the process of evaluating and predicting the performance of gold bonding wire (a material used in semiconductor devices) systems using advanced algorithms and data analysis techniques. Intelligent gold bonding wire system performance prediction can improve the production efficiency and reliability of semiconductor devices.

[0003] Traditional intelligent gold bonding wire system performance prediction usually relies on experienced engineers to build simple physical models based on limited experimental data and physical laws. These models often lack the ability to capture complex data relationships and are difficult to adapt to changes in the production environment, thereby reducing the efficiency and reliability of gold bonding wire performance prediction. Summary of the invention

[0004] The present invention provides a big data driven intelligent gold bonding wire system performance prediction method and system, the main purpose of which is to improve the performance analysis effect of the gold bonding wire.

[0005] To achieve the above object, the present invention provides a big data driven intelligent gold bonding wire system performance prediction method, comprising:

[0006] Obtaining system performance analysis requirements of the smart gold bonding wire system, collecting historical gold bonding wire data of the smart gold bonding wire system according to the system performance analysis requirements, analyzing physical quantities of the gold bonding wires corresponding to the smart gold bonding wire system according to the system performance analysis requirements, and analyzing physical properties of the gold bonding wires based on the physical quantities using the historical gold bonding wire data, wherein the physical properties include material properties, thermodynamic behaviors, and mechanical properties;

[0007] Defining a physical differential equation of the gold bonding wire according to the physical characteristics, constructing a gold wire physical model of the gold bonding wire through the physical differential equation, defining a data-driven network architecture of the gold bonding wire, and determining a data-driven model of the gold bonding wire through the data-driven network architecture;

[0008] Defining a GANs network of the intelligent gold bonding wire system, and analyzing a gold wire feature tuple of the gold bonding wire through the GANs network and the historical gold bonding wire data, and training the gold wire physical model and the data driven model through the gold wire feature tuple to obtain a training gold wire physical model and a training data driven model;

[0009] Analyze the physical model performance and the driving model performance of the training gold wire physical model and the training data driven model, and when the physical model performance and the driving model performance meet the preset physical model performance threshold and the driving model performance, couple the training gold wire physical model and the training data driven model to obtain a hybrid gold wire analysis model;

[0010] Real-time data of the gold wire bonding wire is collected, and a preset attention function is used to capture the performance factor dependency of the gold wire bonding wire through the real-time data of the gold wire. Based on the performance factor dependency and the real-time data of the gold wire, the hybrid gold wire analysis model is used to analyze the potential performance coefficient of the gold wire bonding wire. Based on the potential performance coefficient, a performance analysis report of the gold wire bonding wire is constructed.

[0011] Optionally, analyzing the physical property of the gold bonding wire based on the physical quantity and using the historical gold bonding wire data comprises:

[0012] determining a physical property correlation coefficient of the physical quantity;

[0013] defining a target physical quantity among the physical quantities according to the physical property correlation coefficient;

[0014] Screening the physical quantity gold wire data of the target physical quantity from the historical bonding gold wire data;

[0015] Extracting material property data from the physical quantity gold wire data, and analyzing changes in environmental material properties of the bonding gold wire through the material property data;

[0016] Determining the material property of the bonding gold wire according to the change of the material property of the environment;

[0017] Extracting thermodynamic parameters from the gold wire data of the physical quantity;

[0018] Analyzing the thermodynamic behavior of the bonding gold wire based on the thermodynamic parameters;

[0019] Extracting mechanical parameters from the physical quantity gold wire data;

[0020] Analyzing the mechanical response of the gold bonding wire by using the mechanical parameters, and determining the mechanical properties of the gold bonding wire based on the mechanical response;

[0021] The physical characteristics of the gold bonding wire are analyzed based on the material properties, the thermodynamic behavior, and the mechanical properties.

[0022] Optionally, analyzing the mechanical response of the bonding gold wire by using the mechanical parameters includes: marking the gold wire stress and gold wire strain of the bonding gold wire by using the mechanical parameters;

[0023] Analyzing the crack length of the bonding gold wire under the gold wire stress;

[0024] Analyzing the fracture toughness and strain energy density of the bonding gold wire based on the gold wire stress, the gold wire strain and the crack length;

[0025] dividing the gold wire strain into initial elastic strain and creep strain;

[0026] According to the gold wire stress, the gold wire strain, the initial elastic strain and the creep strain, the creep analysis function of the bonding gold wire is constructed using the following formula, wherein the creep analysis function is:

[0027] ;

[0028] in, represents the wire strain, represents the initial elastic strain, represents the creep strain, represents the wire stress, represents the elastic modulus of the bonding wire, represents the activation energy, is the gas constant, represents absolute temperature, Indicates time, represents the time index, represents the material constant of the bonding wire, represents the exponential function;

[0029] Analyzing the gold wire creep coefficient of the bonding gold wire according to the creep analysis function;

[0030] The mechanical response of the bonding gold wire is determined according to the fracture toughness, the strain energy density and the gold wire creep coefficient.

[0031] Optionally, defining the physical differential equation of the bonding gold wire according to the physical property comprises:

[0032] identifying geometric features of the bonding wire;

[0033] Constructing a geometric model of the bonding gold wire according to the geometric features;

[0034] Based on the physical characteristics, defining the elastic relationship, plastic behavior and temperature influence relationship of the geometric model;

[0035] The physical differential equation of the bonding gold wire is constructed by using the elastic relationship, the plastic behavior, the temperature influence relationship and the gold wire creep coefficient corresponding to the bonding gold wire.

[0036] Optionally, constructing the physical differential equation of the bonding gold wire through the elastic relationship, the plastic behavior, the temperature influence relationship and the gold wire creep coefficient corresponding to the bonding gold wire includes:

[0037] defining a target stress for the bonding wire;

[0038] Analyze the elastic strain, plastic strain, thermal strain and creep strain of the bonding gold wire under the target stress through the elastic relationship, the plastic behavior, the temperature influence relationship and the gold wire creep coefficient corresponding to the bonding gold wire;

[0039] According to the elastic strain, plastic strain, thermal strain, creep strain and target stress, the physical differential equation of the bonding wire is constructed using the following formula:

[0040] ;

[0041] in, represents the target strain of the bonding wire under the target stress, represents the elastic strain of the bonding wire under the target stress, represents the plastic strain of the bonding wire under the target stress, represents the thermal strain of the bonding wire under the target stress, represents the creep strain of the bonding wire under the target stress, represents the differential symbol, Indicates time.

[0042] Optionally, determining the data drive model of the gold bonding wire by using the data drive network architecture includes:

[0043] Cleaning the historical gold bonding wire data of the gold bonding wire to obtain cleaned gold bonding wire data;

[0044] Converting the cleaning bonding gold wire data into gold wire time series data;

[0045] Define a data window for the golden wire time series data;

[0046] Determine a data-driven initial model of the bonding gold wire using a data-driven network architecture according to the data window;

[0047] Using the preset Keras API components, define the model parameters of the data-driven initial model, and add the RNN layer and the fully connected layer of the data-driven initial model;

[0048] The data-driven initial model is updated through the model parameters, the RNN layer and the full-link layer to obtain the data-driven model of the bonding wire.

[0049] Optionally, the GANs network defining the smart gold bonding wire system includes:

[0050] Determine the GANs application target of the smart gold bonding wire system;

[0051] According to the GANs application target, define a GANs generator for the smart gold bonding wire system;

[0052] Constructing a GANs discriminator for the intelligent gold bonding wire system;

[0053] Establishing a shared layer of the GANs generator and the GANs discriminator;

[0054] Adding conditional information of the GANs generator and the GANs discriminator;

[0055] The GANs network of the intelligent gold bonding wire system is established by using the shared layer and the conditional information and utilizing the GANs generator and the GANs discriminator.

[0056] Optionally, capturing the performance factor dependency of the bonding gold wire by using the real-time gold wire data and a preset attention function includes:

[0057] Identify the data complexity coefficient of the gold wire real-time data;

[0058] Defining an attention function parameter of the attention function according to the data complexity coefficient;

[0059] Calculating the performance factor weight of the bonding gold wire using the attention function through the attention function parameters;

[0060] Based on the performance factor weights, a performance factor dependency of the bonding gold wire is determined.

[0061] Optionally, calculating the performance factor weight of the bonding gold wire by using the attention function through the attention function parameter includes:

[0062] According to the attention function parameters, define a weight matrix and a bias term of the attention function;

[0063] Based on the weight matrix and the bias term, the performance factor weight of the bonding wire is calculated using the attention function:

[0064] ;

[0065] in, Indicates the bonding wire The performance factor weight of each performance factor is represents the weight matrix, represents the bias term, represents the attention function, Bonding wire The eigenvector of the performance factors.

[0066] In order to solve the above problems, the present invention also provides a big data driven intelligent gold bonding wire system performance prediction system, the system comprising:

[0067] a physical property analysis module, used to obtain system performance analysis requirements of the intelligent gold bonding wire system, collect historical gold bonding wire data of the intelligent gold bonding wire system according to the system performance analysis requirements, analyze physical quantities of the gold bonding wire corresponding to the intelligent gold bonding wire system according to the system performance analysis requirements, and analyze physical properties of the gold bonding wire based on the physical quantities using the historical gold bonding wire data, wherein the physical properties include material properties, thermodynamic behaviors, and mechanical properties;

[0068] a gold wire model building module, used to define a physical differential equation of the gold bonding wire according to the physical characteristics, and to build a gold wire physical model of the gold bonding wire through the physical differential equation, define a data-driven network architecture of the gold bonding wire, and determine a data-driven model of the gold bonding wire through the data-driven network architecture;

[0069] A gold wire model training module, used to define a GANs network of the intelligent gold wire bonding system, and analyze the gold wire feature tuple of the gold wire bonding through the GANs network and the historical gold wire bonding data, and train the gold wire physical model and the data-driven model through the gold wire feature tuple to obtain a training gold wire physical model and a training data-driven model;

[0070] A gold wire model coupling module is used to analyze the physical model performance and the driving model performance of the training gold wire physical model and the training data driven model. When the physical model performance and the driving model performance meet the preset physical model performance threshold and the driving model performance, the training gold wire physical model and the training data driven model are coupled to obtain a hybrid gold wire analysis model.

[0071] A gold wire performance analysis module is used to collect real-time gold wire data of the bonding gold wire, and capture the performance factor dependency of the bonding gold wire through the gold wire real-time data using a preset attention function; based on the performance factor dependency and the gold wire real-time data, the hybrid gold wire analysis model is used to analyze the potential performance coefficient of the bonding gold wire; based on the potential performance coefficient, a performance analysis report of the bonding gold wire is constructed.

[0072] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0073] at least one processor; and,

[0074] a memory communicatively connected to the at least one processor; wherein,

[0075] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to implement the above-mentioned big data-driven intelligent bonding wire system performance prediction method.

[0076] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned big data-driven intelligent bonding wire system performance prediction method.

[0077] Compared with the problems described in the background technology, the present invention analyzes the physical quantities of the bonding wires corresponding to the intelligent bonding wire system through the system performance analysis requirements, which can more deeply understand the behavior of the bonding wires in the intelligent system, and provide necessary information for establishing an accurate performance prediction model; the present invention defines the physical differential equations of the bonding wires according to the physical properties to provide a basis for constructing a physical model of the gold wire; further, the present invention constructs the gold wire physical model of the bonding wire through the physical differential equation, which can better analyze the physical properties of the bonding wires. Furthermore, the present invention determines through the data-driven network architecture that the data-driven model of the bonding wires can be modeled through a machine learning method, so as to achieve efficient and accurate prediction of the performance of unknown gold wire samples. Furthermore, the present invention defines the GANs network of the intelligent bonding wire system to synthesize data for the bonding wires, thereby improving the training effect of the model and increasing the performance analysis of the bonding wires. Reliability, further, when the physical model performance and the driving model performance meet the preset physical model performance threshold and driving model performance, the present invention couples the training gold wire physical model and the training data driven model to obtain a hybrid gold wire analysis model, which can utilize the advantages of the physical model in explaining physical phenomena, and at the same time utilize the data driven model's ability to process complex data relationships and pattern recognition to improve the model's prediction ability for the performance of the gold bonding wire. Next, the present invention uses the real-time data of the gold wire and the preset attention function to capture the performance factor dependency of the gold bonding wire, which can reveal which performance factors have the most significant impact on the gold wire performance, thereby improving the performance analysis effect of the hybrid gold wire analysis model on the gold bonding wire. Finally, based on the performance factor dependency and the real-time data of the gold wire, the present invention uses the hybrid gold wire analysis model to analyze the potential performance coefficient of the gold bonding wire, which can realize the efficient performance analysis of the gold bonding wire. Therefore, the present invention can improve the performance analysis effect of the gold bonding wire. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 A schematic flow chart of a method for predicting performance of a smart gold bonding wire system driven by big data according to an embodiment of the present invention;

[0079] Figure 2 A functional module diagram of a big data driven intelligent gold bonding wire system performance prediction system provided by an embodiment of the present invention;

[0080] Figure 3 A schematic diagram of the structure of an electronic device of a big data driven intelligent gold bonding wire system performance prediction system provided by an embodiment of the present invention;

[0081] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0082] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0083] The embodiment of the present application provides a big data-driven intelligent bonding wire system performance prediction method. The execution subject of the big data-driven intelligent bonding wire system performance prediction method includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the big data-driven intelligent bonding wire system performance prediction method can be executed by software or hardware installed in a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0084] Embodiment 1:

[0085] Reference Figure 1 FIG. 1 is a flow chart of a method for predicting the performance of a gold bonding wire system driven by big data according to an embodiment of the present invention. In this embodiment, the method for predicting the performance of a gold bonding wire system driven by big data includes:

[0086] S1. Obtain system performance analysis requirements of the intelligent gold bonding wire system, collect historical gold bonding wire data of the intelligent gold bonding wire system according to the system performance analysis requirements, analyze physical quantities of the gold bonding wire corresponding to the intelligent gold bonding wire system according to the system performance analysis requirements, and based on the physical quantities, analyze the physical properties of the gold bonding wire using the historical gold bonding wire data, wherein the physical properties include material properties, thermodynamic behaviors, and mechanical properties.

[0087] It should be explained that the system performance analysis requirements refer to the specific description of the performance standards, indicators and constraints that the intelligent gold bonding wire system needs to meet during operation. These requirements usually include but are not limited to key performance indicators such as system reliability, stability, efficiency, response time, failure rate, prediction accuracy, etc. The historical gold bonding wire data refers to various information and data related to the gold bonding wire system that are monitored, recorded and collected over a period of time in the past. These data usually include parameters such as process parameters, material properties, equipment status and environmental conditions.

[0088] The present invention can analyze the physical quantities of the bonding wires corresponding to the intelligent bonding wire system through the system performance analysis requirements to more deeply understand the behavior of the bonding wires in the intelligent system and provide necessary information for establishing an accurate performance prediction model. In detail, the physical quantities refer to those quantitative indicators that can directly or indirectly reflect the performance of the bonding wires.

[0089] Optionally, the present invention may analyze the physical characteristics of the bonding wire based on the physical quantity and using the historical bonding wire data to form a bonding wire physical property analysis report.

[0090] In detail, the analyzing the physical properties of the gold bonding wire based on the physical quantity and using the historical gold bonding wire data includes:

[0091] determining a physical property correlation coefficient of the physical quantity;

[0092] defining a target physical quantity among the physical quantities according to the physical property correlation coefficient;

[0093] Screening the physical quantity gold wire data of the target physical quantity from the historical bonding gold wire data;

[0094] Extracting material property data from the physical quantity gold wire data, and analyzing changes in environmental material properties of the bonding gold wire through the material property data;

[0095] Determining the material property of the bonding gold wire according to the change of the material property of the environment;

[0096] Extracting thermodynamic parameters from the gold wire data of the physical quantity;

[0097] Analyzing the thermodynamic behavior of the bonding gold wire based on the thermodynamic parameters;

[0098] Extracting mechanical parameters from the physical quantity gold wire data;

[0099] Analyzing the mechanical response of the gold bonding wire by using the mechanical parameters, and determining the mechanical properties of the gold bonding wire based on the mechanical response;

[0100] The physical characteristics of the gold bonding wire are analyzed based on the material properties, the thermodynamic behavior, and the mechanical properties.

[0101] Among them, the physical property correlation coefficient refers to a statistical indicator that measures the degree of correlation between different physical quantities and the physical properties of the bonding wire (such as material properties, thermodynamic behavior, and mechanical properties); the target physical quantity refers to the physical quantity selected as the most representative for analyzing the physical properties of the bonding wire based on the physical property correlation coefficient; the physical quantity gold wire data refers to a data set related to the target physical quantity screened out from the historical bonding wire data; the material property data refers to data related to the material properties of the bonding wire extracted from the physical quantity gold wire data, such as elastic modulus, yield strength, etc.; the environmental material property change refers to the change in the material properties of the bonding wire caused by environmental factors (such as temperature, humidity, chemical corrosion, etc.) identified after analyzing the material property data; the material property refers to the inherent properties of the bonding wire, such as elastic modulus. The properties of the bonding wire are as follows: thermal conductivity, specific heat capacity, thermal expansion coefficient, thermal conductivity, etc., which determine the behavior of the material under conditions such as force and heat. The thermodynamic parameters refer to the physical quantities that describe the thermal behavior of the bonding wire, such as thermal expansion coefficient, thermal conductivity, specific heat capacity and other parameters. The thermodynamic behavior refers to the behavior of the bonding wire under thermal action obtained based on the analysis of thermodynamic parameters, such as thermal expansion, heat conduction and other behaviors. The mechanical parameters refer to the physical quantities that describe the mechanical behavior of the bonding wire, such as stress, strain, bonding force and other parameters. The mechanical response refers to the reaction of the bonding wire when subjected to mechanical action, such as deformation, fracture, etc. The mechanical properties refer to the mechanical properties of the bonding wire, such as strength, toughness, fatigue life and other properties. The physical properties refer to the properties that integrate material properties, thermodynamic behavior and mechanical properties. They together describe the overall performance of the bonding wire in a specific application environment.

[0102] Further, analyzing the mechanical response of the bonding gold wire by the mechanical parameters includes: marking the gold wire stress and gold wire strain of the bonding gold wire by the mechanical parameters;

[0103] Analyzing the crack length of the bonding gold wire under the gold wire stress;

[0104] Analyzing the fracture toughness and strain energy density of the bonding gold wire based on the gold wire stress, the gold wire strain and the crack length;

[0105] dividing the gold wire strain into initial elastic strain and creep strain;

[0106] According to the gold wire stress, the gold wire strain, the initial elastic strain and the creep strain, the creep analysis function of the bonding gold wire is constructed using the following formula, wherein the creep analysis function is:

[0107] ;

[0108] in, represents the wire strain, represents the initial elastic strain, represents the creep strain, represents the wire stress, represents the elastic modulus of the bonding wire, represents the activation energy, is the gas constant, represents absolute temperature, Indicates time, represents the time index, represents the material constant of the bonding wire, represents the exponential function;

[0109] Analyzing the gold wire creep coefficient of the bonding gold wire according to the creep analysis function;

[0110] The mechanical response of the bonding gold wire is determined according to the fracture toughness, the strain energy density and the gold wire creep coefficient.

[0111] Among them, the crack length refers to the length of the crack, the fracture toughness refers to the ability of the material to resist crack extension, the strain energy density refers to the energy density inside the material, the creep analysis function refers to a mathematical function that describes creep behavior, the elastic modulus refers to the ability of the material to resist deformation, the activation energy refers to the energy required to start the creep process, the gas constant refers to a constant in physics, the gold wire stress refers to the stress acting on the gold wire, the gold wire strain refers to the strain of the gold wire, the absolute temperature refers to the temperature of the material, the time index refers to an index that describes the effect of time on creep, and the material constant refers to a constant specific to the material.

[0112] S2. Define the physical differential equation of the bonding gold wire according to the physical characteristics, and construct a gold wire physical model of the bonding gold wire through the physical differential equation, define the data-driven network architecture of the bonding gold wire, and determine the data-driven model of the bonding gold wire through the data-driven network architecture.

[0113] The present invention defines the physical differential equation of the bonding gold wire according to the physical characteristics to provide a basis for constructing a physical model of the gold wire.

[0114] In detail, the physical differential equation of the bonding wire is defined according to the physical property, including:

[0115] identifying geometric features of the bonding wire;

[0116] Constructing a geometric model of the bonding gold wire according to the geometric features;

[0117] Based on the physical characteristics, defining the elastic relationship, plastic behavior and temperature influence relationship of the geometric model;

[0118] The physical differential equation of the bonding gold wire is constructed by using the elastic relationship, the plastic behavior, the temperature influence relationship and the gold wire creep coefficient corresponding to the bonding gold wire.

[0119] Among them, the geometric features refer to quantifiable properties such as the shape, size and structure of the bonding gold wire. The geometric model is a mathematical representation established based on the geometric features, and is a model used to describe the shape and size of the bonding gold wire. The elastic relationship refers to a description of the relationship between stress and strain within the elastic range. The plastic behavior refers to a description of how the material undergoes permanent deformation after the stress exceeds the yield point. The temperature influence relationship refers to a description of the effect of temperature changes on material properties and mechanical behavior. The physical differential equation refers to a mathematical equation that is established based on the above-mentioned geometric model and physical properties to describe the mechanical behavior of the gold wire.

[0120] Furthermore, the physical differential equation of the bonding gold wire is constructed by using the elastic relationship, the plastic behavior, the temperature influence relationship and the gold wire creep coefficient corresponding to the bonding gold wire, including:

[0121] defining a target stress for the bonding wire;

[0122] Analyze the elastic strain, plastic strain, thermal strain and creep strain of the bonding gold wire under the target stress through the elastic relationship, the plastic behavior, the temperature influence relationship and the gold wire creep coefficient corresponding to the bonding gold wire;

[0123] According to the elastic strain, plastic strain, thermal strain, creep strain and target stress, the physical differential equation of the bonding wire is constructed using the following formula:

[0124] ;

[0125] in, represents the target strain of the bonding wire under the target stress, represents the elastic strain of the bonding wire under the target stress, represents the plastic strain of the bonding wire under the target stress, represents the thermal strain of the bonding wire under the target stress, represents the creep strain of the bonding wire under the target stress, represents the differential symbol, Indicates time.

[0126] Among them, the target stress refers to the expected stress level applied to the bonding gold wire, the elastic strain refers to the reversible strain of the bonding gold wire under the target stress, the plastic strain refers to the irreversible strain of the bonding gold wire under the target stress, the thermal strain refers to the strain of the bonding gold wire caused by temperature change, the creep strain refers to the strain gradually occurring in the bonding gold wire over time under the long-term action of the target stress, and the target strain refers to the total strain of the bonding gold wire under the target stress, which is the sum of elastic strain, plastic strain, thermal strain and creep strain.

[0127] The present invention can better analyze the physical properties of the bonding gold wire by constructing the gold wire physical model of the bonding gold wire through the physical differential equation. The gold wire physical model refers to a model that describes the physical properties of the bonding gold wire. In detail, the gold wire physical model is constructed by defining constraint conditions (initial stress, strain and temperature) in the physical differential equation.

[0128] Optionally, the data-driven network architecture of the bonding wire defined in the present invention can be used as a basis for constructing a data-driven model of the bonding wire. In detail, the data-driven network architecture refers to a data-based network architecture design method that focuses on learning features and patterns from data rather than relying on pre-set rules. The data-driven network architecture can be designed by a recurrent neural network.

[0129] The present invention uses the data-driven network architecture to determine that the data-driven model of the bonding gold wire can be modeled through a machine learning method, thereby achieving efficient and accurate prediction of the performance of unknown gold wire samples.

[0130] In detail, determining the data drive model of the gold bonding wire through the data drive network architecture includes:

[0131] Cleaning the historical gold bonding wire data of the gold bonding wire to obtain cleaned gold bonding wire data;

[0132] Converting the cleaning bonding gold wire data into gold wire time series data;

[0133] Define a data window for the golden wire time series data;

[0134] Determine a data-driven initial model of the bonding gold wire using a data-driven network architecture according to the data window;

[0135] Using the preset Keras API components, define the model parameters of the data-driven initial model, and add the RNN layer and the fully connected layer of the data-driven initial model;

[0136] The data-driven initial model is updated through the model parameters, the RNN layer and the full-link layer to obtain the data-driven model of the bonding wire.

[0137] Among them, the cleaning of bonding gold wire data refers to processing the original collected bonding gold wire data to eliminate errors, outliers, duplicate records, incomplete entries or any data points that do not meet the analysis requirements. The gold wire time series data refers to organizing the cleaned bonding gold wire data into a sequence of data points arranged in chronological order. The data window refers to a continuous data segment of a fixed length extracted from the time series data, which is used as the input of the model. The data-driven initial model refers to a preliminary model structure defined based on the data window. The Keras API component refers to a component of various predefined neural network layers, optimizers, loss functions, activation functions and other parameters provided by Keras. The model parameters refer to variables that define the neural network structure and training process, including but not limited to the number of layers, the number of neurons in each layer, the learning rate, the batch size, the number of iterations, the regularization parameter and other parameters. The RNN layer refers to a layer in a recurrent neural network, which can process sequence data and affect subsequent outputs by memorizing previous information. The fully connected layer refers to a standard layer in a neural network, in which each neuron is connected to all neurons in the previous layer. The data-driven model refers to the final model after adjustment.

[0138] S3. Define the GANs network of the intelligent gold wire bonding system, and analyze the gold wire feature tuple of the gold wire bonding wire through the GANs network and the historical gold wire bonding data, and train the gold wire physical model and the data-driven model through the gold wire feature tuple to obtain the trained gold wire physical model and the trained data-driven model.

[0139] The present invention defines that the GANs network of the intelligent gold bonding wire system can synthesize data on the gold bonding wires to improve the training effect of the model and increase the reliability of the performance analysis of the gold bonding wires.

[0140] In detail, the GANs network defining the smart gold bonding wire system includes:

[0141] Determine the GANs application target of the smart gold bonding wire system;

[0142] According to the GANs application target, define a GANs generator for the smart gold bonding wire system;

[0143] Constructing a GANs discriminator for the intelligent gold bonding wire system;

[0144] Establishing a shared layer of the GANs generator and the GANs discriminator;

[0145] Adding conditional information of the GANs generator and the GANs discriminator;

[0146] The GANs network of the intelligent gold bonding wire system is established by using the shared layer and the conditional information and utilizing the GANs generator and the GANs discriminator.

[0147] The GANs generator refers to a neural network whose purpose is to receive a random noise vector (usually drawn from a simple distribution, such as a normal distribution) and generate data samples similar to real data. The GANs discriminator refers to another neural network whose task is to receive the output from the generator and the data of the real data set and determine whether the data is real data or fake data generated by the generator. In the bonding wire system, the GANs discriminator evaluates the authenticity of the data samples produced by the generator. The shared layer refers to the neural network layer shared between the GANs generator and the GANs discriminator. These layers can include convolutional layers, fully connected layers, or other types of layers. The conditional information refers to additional information added to the input of the GANs generator and the GANs discriminator, which guides the GANs generator to generate data samples with specific attributes or characteristics. The GANs network refers to the overall model composed of the GANs generator and the GANs discriminator.

[0148] Optionally, the condition information for adding the GANs generator and the GANs discriminator may be parameters such as the type of gold wire, manufacturing parameters, and expected physical properties.

[0149] By analyzing the gold wire feature tuple of the gold wire through the GANs network and the historical gold wire bonding data, a large number of gold wire feature tuples can be generated through the GANs network to train the model, thereby obtaining an in-depth understanding of the behavior of the gold wire, thereby improving the accuracy of the gold wire bonding analysis. In detail, the gold wire feature tuple refers to a set of attributes that describe the characteristics of gold wire (usually used for bonding wires in semiconductor device connections).

[0150] It should be explained that the training gold wire physical model refers to a model obtained by training the gold wire physical model through the gold wire feature tuple, and the training data-driven model refers to a model obtained by training the data-driven model through the gold wire feature tuple.

[0151] S4, analyzing the physical model performance and the driving model performance of the training gold wire physical model and the training data driven model, and when the physical model performance and the driving model performance meet the preset physical model performance threshold and the driving model performance, coupling the training gold wire physical model and the training data driven model to obtain a hybrid gold wire analysis model;

[0152] It should be explained that the physical model performance refers to the performance metric of the trained gold wire physical model when completing a specific task, and the driving model performance refers to the performance metric of the trained data driven model when completing a specific task.

[0153] In the present invention, when the physical model performance and the driving model performance meet the preset physical model performance threshold and driving model performance, the training gold wire physical model and the training data driven model are coupled to obtain a hybrid gold wire analysis model, which can utilize the advantages of the physical model in explaining physical phenomena, and at the same time utilize the data driven model's ability to process complex data relationships and pattern recognition, thereby improving the model's ability to predict the performance of the bonding gold wire. The hybrid gold wire analysis model refers to an analysis model that combines a traditional physical model with a data driven model.

[0154] S5. Collect the real-time data of the gold wire of the bonding gold wire, and use a preset attention function to capture the performance factor dependency of the bonding gold wire through the real-time data of the gold wire. Based on the performance factor dependency and the real-time data of the gold wire, use the hybrid gold wire analysis model to analyze the potential performance coefficient of the bonding gold wire, and based on the potential performance coefficient, construct a performance analysis report of the bonding gold wire.

[0155] It should be explained that the real-time data of the gold wire refers to the data obtained by continuously monitoring the physical state, performance parameters or environmental conditions of the gold wire during the production process or actual application, such as temperature data, mechanical property data, electrical performance data, process parameters, visual monitoring data and environmental conditions.

[0156] The present invention uses the real-time gold wire data and a preset attention function to capture the performance factor dependencies of the bonding gold wire, which can reveal which performance factors have the most significant impact on the gold wire performance, thereby improving the effect of the hybrid gold wire analysis model on the performance analysis of the bonding gold wire.

[0157] In detail, the real-time data of the gold wire is used to capture the performance factor dependency of the bonding gold wire using a preset attention function, including:

[0158] Identify the data complexity coefficient of the gold wire real-time data;

[0159] Defining an attention function parameter of the attention function according to the data complexity coefficient;

[0160] Calculating the performance factor weight of the bonding gold wire using the attention function through the attention function parameters;

[0161] Based on the performance factor weights, a performance factor dependency of the bonding gold wire is determined.

[0162] Among them, the data complexity coefficient refers to one or more indicators used to quantify the complexity of the data set, the attention function parameters refer to the parameters used to configure and optimize the attention mechanism, the performance factor weight refers to the value calculated by the attention mechanism, which represents the degree of influence of each performance factor on the model output, and the performance factor dependency refers to the relationship between different performance factors and their degree of influence on the output results.

[0163] Further, the calculating the performance factor weight of the bonding gold wire by using the attention function through the attention function parameter includes:

[0164] According to the attention function parameters, define a weight matrix and a bias term of the attention function;

[0165] Based on the weight matrix and the bias term, the performance factor weight of the bonding wire is calculated using the attention function:

[0166] ;

[0167] in, Indicates the bonding wire The performance factor weight of each performance factor is represents the weight matrix, represents the bias term, represents the attention function, Bonding wire The eigenvector of the performance factors.

[0168] Among them, the weight matrix refers to a learnable parameter matrix, the bias term refers to a learnable parameter vector that can be added to the attention score to provide additional flexibility, the performance factor weight refers to the relative importance of the performance factor among all performance factors, the attention function refers to the attention weight calculated based on the query and key vectors, and the feature vector refers to the feature representation of the performance factor, which is usually a numerical vector.

[0169] The present invention can realize efficient performance analysis of the bonding wire by analyzing the potential performance coefficient of the bonding wire based on the performance factor dependency and the real-time data of the bonding wire using the hybrid bonding wire analysis model, wherein the potential performance coefficient refers to a key indicator affecting the performance of the bonding wire, which is usually related to the physical properties, chemical properties, mechanical properties of the bonding wire and its performance in a specific application. The performance analysis report refers to a report on the performance analysis of the bonding wire, including performance analysis, performance prediction failure, performance prediction failure level, performance prediction failure optimization method and the like.

[0170] Compared with the problems described in the background technology, the present invention analyzes the physical quantities of the bonding wires corresponding to the intelligent bonding wire system through the system performance analysis requirements, which can more deeply understand the behavior of the bonding wires in the intelligent system, and provide necessary information for establishing an accurate performance prediction model; the present invention defines the physical differential equations of the bonding wires according to the physical properties to provide a basis for constructing a physical model of the gold wire; further, the present invention constructs the gold wire physical model of the bonding wire through the physical differential equation, which can better analyze the physical properties of the bonding wires. Furthermore, the present invention determines through the data-driven network architecture that the data-driven model of the bonding wires can be modeled through a machine learning method, so as to achieve efficient and accurate prediction of the performance of unknown gold wire samples. Furthermore, the present invention defines the GANs network of the intelligent bonding wire system to synthesize data for the bonding wires, thereby improving the training effect of the model and increasing the performance analysis of the bonding wires. Reliability, further, when the physical model performance and the driving model performance meet the preset physical model performance threshold and driving model performance, the present invention couples the training gold wire physical model and the training data driven model to obtain a hybrid gold wire analysis model, which can utilize the advantages of the physical model in explaining physical phenomena, and at the same time utilize the data driven model's ability to process complex data relationships and pattern recognition to improve the model's prediction ability for the performance of the gold bonding wire. Next, the present invention uses the real-time data of the gold wire and the preset attention function to capture the performance factor dependency of the gold bonding wire, which can reveal which performance factors have the most significant impact on the gold wire performance, thereby improving the performance analysis effect of the hybrid gold wire analysis model on the gold bonding wire. Finally, based on the performance factor dependency and the real-time data of the gold wire, the present invention uses the hybrid gold wire analysis model to analyze the potential performance coefficient of the gold bonding wire, which can realize the efficient performance analysis of the gold bonding wire. Therefore, the present invention can improve the performance analysis effect of the gold bonding wire.

[0171] Embodiment 2:

[0172] like Figure 2 , which is a functional module diagram of a big data driven intelligent gold bonding wire system performance prediction system provided by an embodiment of the present invention.

[0173] The big data driven intelligent gold bonding wire system performance prediction system 200 of the present invention can be installed in an electronic device. According to the functions implemented, the big data driven intelligent gold bonding wire system performance prediction system 200 can include a physical property analysis module 201, a gold wire model construction module 202, a gold wire model training module 203, a gold wire model coupling module 204 and a gold wire performance analysis module 205. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0174] In this embodiment, the functions of each module / unit are as follows:

[0175] The physical property analysis module 201 is used to obtain system performance analysis requirements of the smart gold bonding wire system, collect historical gold bonding wire data of the smart gold bonding wire system according to the system performance analysis requirements, analyze physical quantities of the gold bonding wires corresponding to the smart gold bonding wire system according to the system performance analysis requirements, and analyze physical properties of the gold bonding wires based on the physical quantities using the historical gold bonding wire data, wherein the physical properties include material properties, thermodynamic behaviors, and mechanical properties;

[0176] The gold wire model building module 202 is used to define the physical differential equation of the gold bonding wire according to the physical characteristics, and build the gold wire physical model of the gold bonding wire through the physical differential equation, define the data driven network architecture of the gold bonding wire, and determine the data driven model of the gold bonding wire through the data driven network architecture;

[0177] The gold wire model training module 203 is used to define a GANs network of the intelligent gold bonding wire system, and analyze the gold wire feature tuple of the gold bonding wire through the GANs network and the historical gold bonding wire data, and train the gold wire physical model and the data driven model through the gold wire feature tuple to obtain a training gold wire physical model and a training data driven model;

[0178] The gold wire model coupling module 204 is used to analyze the physical model performance and driving model performance of the training gold wire physical model and the training data driven model, and when the physical model performance and driving model performance meet the preset physical model performance threshold and driving model performance, the training gold wire physical model and the training data driven model are coupled to obtain a hybrid gold wire analysis model;

[0179] The gold wire performance analysis module 205 is used to collect the gold wire real-time data of the bonding gold wire, and use a preset attention function to capture the performance factor dependency of the bonding gold wire through the gold wire real-time data; based on the performance factor dependency and the gold wire real-time data, use the hybrid gold wire analysis model to analyze the potential performance coefficient of the bonding gold wire; and based on the potential performance coefficient, construct a performance analysis report of the bonding gold wire.

[0180] In detail, the modules described in the big data driven intelligent bonding wire system performance prediction system 200 described in the embodiment of the present invention adopt the same technical means as the big data driven intelligent bonding wire system performance prediction method described in the accompanying drawings when used, and can produce the same technical effects, which will not be repeated here.

[0181] An embodiment of the present invention provides an electronic device for implementing a big data driven intelligent gold bonding wire system performance prediction method.

[0182] See also Figure 3 As shown, the electronic device may include a processor 30, a memory 31, a communication bus 32 and a communication interface 33, and may also include a computer program stored in the memory 31 and executable on the processor 30, such as a big data driven intelligent bonding wire system performance prediction method program.

[0183] In some embodiments, the processor may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and a combination of various control chips, etc. The processor is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes various functions of the electronic device and processes data by running or executing programs or modules stored in the memory (for example, executing a big data-driven intelligent bonding wire system performance prediction program, etc.), and calling data stored in the memory.

[0184] The memory includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the memory may be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory may also be an external storage device of an electronic device, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Furthermore, the memory may also include both an internal storage unit and an external storage device of the electronic device. The memory can not only be used to store application software and various types of data installed in the electronic device, such as the code of a smart bonding wire system performance prediction program driven by big data, but also can be used to temporarily store data that has been output or is to be output.

[0185] The communication bus may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory and at least one processor, etc.

[0186] The communication interface is used for communication between the above-mentioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a display (Display), an input unit (such as a keyboard (Keyboard)), and optionally, the user interface may also be a standard wired interface, a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device and to display a visual user interface.

[0187] For example, although not shown, the electronic device may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor through a power management system, so that the power management system can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators. The electronic device may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0188] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0189] The big data driven intelligent gold bonding wire system performance prediction program stored in the memory of the electronic device is a combination of multiple instructions, and when running in the processor, it can achieve:

[0190] Obtaining system performance analysis requirements of the smart gold bonding wire system, collecting historical gold bonding wire data of the smart gold bonding wire system according to the system performance analysis requirements, analyzing physical quantities of the gold bonding wires corresponding to the smart gold bonding wire system according to the system performance analysis requirements, and analyzing physical properties of the gold bonding wires based on the physical quantities using the historical gold bonding wire data, wherein the physical properties include material properties, thermodynamic behaviors, and mechanical properties;

[0191] Defining a physical differential equation of the gold bonding wire according to the physical characteristics, constructing a gold wire physical model of the gold bonding wire through the physical differential equation, defining a data-driven network architecture of the gold bonding wire, and determining a data-driven model of the gold bonding wire through the data-driven network architecture;

[0192] Defining a GANs network of the intelligent gold bonding wire system, and analyzing a gold wire feature tuple of the gold bonding wire through the GANs network and the historical gold bonding wire data, and training the gold wire physical model and the data driven model through the gold wire feature tuple to obtain a training gold wire physical model and a training data driven model;

[0193] Analyze the physical model performance and the driving model performance of the training gold wire physical model and the training data driven model, and when the physical model performance and the driving model performance meet the preset physical model performance threshold and the driving model performance, couple the training gold wire physical model and the training data driven model to obtain a hybrid gold wire analysis model;

[0194] Real-time data of the gold wire bonding wire is collected, and a preset attention function is used to capture the performance factor dependency of the gold wire bonding wire through the real-time data of the gold wire. Based on the performance factor dependency and the real-time data of the gold wire, the hybrid gold wire analysis model is used to analyze the potential performance coefficient of the gold wire bonding wire. Based on the potential performance coefficient, a performance analysis report of the gold wire bonding wire is constructed.

[0195] Specifically, the specific implementation method of the processor for the above instructions can refer to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, which will not be repeated here.

[0196] Furthermore, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or system that can carry the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0197] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0198] Obtaining system performance analysis requirements of the smart gold bonding wire system, collecting historical gold bonding wire data of the smart gold bonding wire system according to the system performance analysis requirements, analyzing physical quantities of the gold bonding wires corresponding to the smart gold bonding wire system according to the system performance analysis requirements, and analyzing physical properties of the gold bonding wires based on the physical quantities using the historical gold bonding wire data, wherein the physical properties include material properties, thermodynamic behaviors, and mechanical properties;

[0199] Defining a physical differential equation of the gold bonding wire according to the physical characteristics, constructing a gold wire physical model of the gold bonding wire through the physical differential equation, defining a data-driven network architecture of the gold bonding wire, and determining a data-driven model of the gold bonding wire through the data-driven network architecture;

[0200] Defining a GANs network of the intelligent gold bonding wire system, and analyzing a gold wire feature tuple of the gold bonding wire through the GANs network and the historical gold bonding wire data, and training the gold wire physical model and the data driven model through the gold wire feature tuple to obtain a training gold wire physical model and a training data driven model;

[0201] Analyze the physical model performance and the driving model performance of the training gold wire physical model and the training data driven model, and when the physical model performance and the driving model performance meet the preset physical model performance threshold and the driving model performance, couple the training gold wire physical model and the training data driven model to obtain a hybrid gold wire analysis model;

[0202] Real-time data of the gold wire bonding wire is collected, and a preset attention function is used to capture the performance factor dependency of the gold wire bonding wire through the real-time data of the gold wire. Based on the performance factor dependency and the real-time data of the gold wire, the hybrid gold wire analysis model is used to analyze the potential performance coefficient of the gold wire bonding wire. Based on the potential performance coefficient, a performance analysis report of the gold wire bonding wire is constructed.

[0203] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0204] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0205] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0206] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0207] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0208] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0209] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems stated in a system claim can also be implemented by one unit or system through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

Claims

1. A big data driven intelligent gold bonding wire system performance prediction method, characterized in that: The method comprises: Obtaining system performance analysis requirements of the smart gold bonding wire system, collecting historical gold bonding wire data of the smart gold bonding wire system according to the system performance analysis requirements, analyzing physical quantities of the gold bonding wires corresponding to the smart gold bonding wire system according to the system performance analysis requirements, and analyzing physical properties of the gold bonding wires based on the physical quantities using the historical gold bonding wire data, wherein the physical properties include material properties, thermodynamic behaviors, and mechanical properties; A physical differential equation of the gold bonding wire is defined according to the physical characteristics, and a gold wire physical model of the gold bonding wire is constructed through the physical differential equation, a data-driven network architecture of the gold bonding wire is defined, and a data-driven model of the gold bonding wire is determined through the data-driven network architecture, wherein the physical differential equation of the gold bonding wire is defined according to the physical characteristics, including: identifying geometric features of the gold bonding wire, constructing a geometric model of the gold bonding wire according to the geometric features, defining an elastic relationship, a plastic behavior, and a temperature influence relationship of the geometric model based on the physical characteristics, and constructing a physical differential equation of the gold bonding wire through the elastic relationship, the plastic behavior, the temperature influence relationship, and a gold wire creep coefficient corresponding to the gold bonding wire, wherein the physical differential equation of the gold bonding wire is constructed through the elastic relationship, the plastic behavior, the temperature influence relationship, and a gold wire creep coefficient corresponding to the gold bonding wire, including: The target stress of the bonding gold wire is defined, and the elastic strain, plastic strain, thermal strain and creep strain of the bonding gold wire under the target stress are analyzed through the elastic relationship, the plastic behavior, the temperature influence relationship and the gold wire creep coefficient corresponding to the bonding gold wire. According to the elastic strain, plastic strain, thermal strain, creep strain and target stress, the physical differential equation of the bonding gold wire is constructed using the following formula: ； in, represents the target strain of the bonding wire under the target stress, represents the elastic strain of the bonding wire under the target stress, represents the plastic strain of the bonding wire under the target stress, represents the thermal strain of the bonding wire under the target stress, represents the creep strain of the bonding wire under the target stress, represents the differential symbol, Indicates time; Defining a GANs network of the intelligent gold bonding wire system, and analyzing a gold wire feature tuple of the gold bonding wire through the GANs network and the historical gold bonding wire data, and training the gold wire physical model and the data driven model through the gold wire feature tuple to obtain a training gold wire physical model and a training data driven model; Analyze the physical model performance and the driving model performance of the training gold wire physical model and the training data driven model, and when the physical model performance and the driving model performance meet the preset physical model performance threshold and the driving model performance, couple the training gold wire physical model and the training data driven model to obtain a hybrid gold wire analysis model; Real-time data of the gold wire bonding wire is collected, and a preset attention function is used to capture the performance factor dependency of the gold wire bonding wire through the real-time data of the gold wire. Based on the performance factor dependency and the real-time data of the gold wire, the hybrid gold wire analysis model is used to analyze the potential performance coefficient of the gold wire bonding wire. Based on the potential performance coefficient, a performance analysis report of the gold wire bonding wire is constructed.

2. The big data driven intelligent gold bonding wire system performance prediction method according to claim 1, characterized in that: The analyzing the physical property of the gold bonding wire by using the historical gold bonding wire data based on the physical quantity comprises: determining a physical property correlation coefficient of the physical quantity; defining a target physical quantity among the physical quantities according to the physical property correlation coefficient; Screening the physical quantity gold wire data of the target physical quantity from the historical bonding gold wire data; Extracting material property data from the physical quantity gold wire data, and analyzing changes in environmental material properties of the bonding gold wire through the material property data; Determining the material property of the bonding gold wire according to the change of the material property of the environment; Extracting thermodynamic parameters from the gold wire data of the physical quantity; Analyzing the thermodynamic behavior of the bonding gold wire based on the thermodynamic parameters; Extracting mechanical parameters from the physical quantity gold wire data; Analyzing the mechanical response of the gold bonding wire by using the mechanical parameters, and determining the mechanical properties of the gold bonding wire based on the mechanical response; The physical characteristics of the gold bonding wire are analyzed based on the material properties, the thermodynamic behavior, and the mechanical properties.

3. The big data driven intelligent gold bonding wire system performance prediction method according to claim 2, characterized in that: Analyzing the mechanical response of the bonding gold wire by using the mechanical parameters includes: marking the gold wire stress and gold wire strain of the bonding gold wire by using the mechanical parameters; Analyzing the crack length of the bonding gold wire under the gold wire stress; Analyzing the fracture toughness and strain energy density of the bonding gold wire based on the gold wire stress, the gold wire strain and the crack length; dividing the gold wire strain into initial elastic strain and creep strain; According to the gold wire stress, the gold wire strain, the initial elastic strain and the creep strain, the creep analysis function of the bonding gold wire is constructed using the following formula, wherein the creep analysis function is: ； in, represents the wire strain, represents the initial elastic strain, represents the creep strain, represents the wire stress, represents the elastic modulus of the bonding wire, represents the activation energy, is the gas constant, represents absolute temperature, Indicates time, represents the time index, represents the material constant of the bonding wire, represents the exponential function; Analyzing the gold wire creep coefficient of the bonding gold wire according to the creep analysis function; The mechanical response of the bonding gold wire is determined according to the fracture toughness, the strain energy density and the gold wire creep coefficient.

4. The big data driven intelligent gold bonding wire system performance prediction method according to claim 3, characterized in that: Determining the data drive model of the bonding gold wire by using the data drive network architecture includes: Cleaning the historical gold bonding wire data of the gold bonding wire to obtain cleaned gold bonding wire data; Converting the cleaning bonding gold wire data into gold wire time series data; Define a data window for the golden wire time series data; Determine a data-driven initial model of the bonding gold wire using a data-driven network architecture according to the data window; Using the preset Keras API components, define the model parameters of the data-driven initial model, and add the RNN layer and the fully connected layer of the data-driven initial model; The data-driven initial model is updated through the model parameters, the RNN layer and the full-link layer to obtain the data-driven model of the bonding wire.

5. The big data driven intelligent gold bonding wire system performance prediction method according to claim 4, characterized in that: The GANs network defining the intelligent gold bonding wire system comprises: Determine the GANs application target of the smart gold bonding wire system; According to the GANs application target, define a GANs generator for the smart gold bonding wire system; Constructing a GANs discriminator for the intelligent gold bonding wire system; Establishing a shared layer of the GANs generator and the GANs discriminator; Adding conditional information of the GANs generator and the GANs discriminator; The GANs network of the intelligent gold bonding wire system is established by using the shared layer and the conditional information and utilizing the GANs generator and the GANs discriminator.

6. The big data driven intelligent gold bonding wire system performance prediction method according to claim 5, characterized in that: The method of capturing the performance factor dependency of the bonding gold wire by using the real-time gold wire data and a preset attention function includes: Identify the data complexity coefficient of the gold wire real-time data; Defining an attention function parameter of the attention function according to the data complexity coefficient; Calculating the performance factor weight of the bonding gold wire using the attention function through the attention function parameters; Based on the performance factor weights, a performance factor dependency of the bonding gold wire is determined.

7. The big data driven intelligent gold bonding wire system performance prediction method according to claim 6, characterized in that: The step of calculating the performance factor weight of the gold bonding wire by using the attention function through the attention function parameter includes: According to the attention function parameters, define a weight matrix and a bias term of the attention function; Based on the weight matrix and the bias term, the performance factor weight of the bonding wire is calculated using the attention function: ； in, Indicates the bonding wire The performance factor weight of each performance factor is represents the weight matrix, represents the bias term, represents the attention function, Bonding wire The eigenvector of the performance factors.

8. A big data driven intelligent gold bonding wire system performance prediction system, characterized in that: A method for predicting the performance of an intelligent gold bonding wire system driven by big data as claimed in any one of claims 1 to 7, the system comprising: a physical property analysis module, used to obtain system performance analysis requirements of the intelligent gold bonding wire system, collect historical gold bonding wire data of the intelligent gold bonding wire system according to the system performance analysis requirements, analyze physical quantities of the gold bonding wire corresponding to the intelligent gold bonding wire system according to the system performance analysis requirements, and analyze physical properties of the gold bonding wire based on the physical quantities using the historical gold bonding wire data, wherein the physical properties include material properties, thermodynamic behaviors, and mechanical properties; a gold wire model building module, used to define a physical differential equation of the gold bonding wire according to the physical characteristics, and to build a gold wire physical model of the gold bonding wire through the physical differential equation, define a data-driven network architecture of the gold bonding wire, and determine a data-driven model of the gold bonding wire through the data-driven network architecture; A gold wire model training module, used to define a GANs network of the intelligent gold wire bonding system, and analyze the gold wire feature tuple of the gold wire bonding through the GANs network and the historical gold wire bonding data, and train the gold wire physical model and the data-driven model through the gold wire feature tuple to obtain a training gold wire physical model and a training data-driven model; A gold wire model coupling module is used to analyze the physical model performance and the driving model performance of the training gold wire physical model and the training data driven model. When the physical model performance and the driving model performance meet the preset physical model performance threshold and the driving model performance, the training gold wire physical model and the training data driven model are coupled to obtain a hybrid gold wire analysis model. A gold wire performance analysis module is used to collect real-time gold wire data of the bonding gold wire, and capture the performance factor dependency of the bonding gold wire through the gold wire real-time data using a preset attention function; based on the performance factor dependency and the gold wire real-time data, the hybrid gold wire analysis model is used to analyze the potential performance coefficient of the bonding gold wire; based on the potential performance coefficient, a performance analysis report of the bonding gold wire is constructed.

Images