Bonding silver wire service life prediction method and system based on big data

Abstract

The invention provides a bonding silver wire service life prediction method and system based on big data. The method comprises the following steps: establishing a welding spot combination service life model based on data information of a plurality of bonding welding spots; obtaining a bonding welding spot scanning image of the bonding welding spot through a scanning electron microscope; analyzing the bonding welding spot scanning image to obtain welding spot combination data; and inputting the welding spot bonding data into the welding spot bonding life model to obtain the predicted bonding service life of the bonding welding spot, so that the service life of the bonding welding spot can be accurately estimated, and the problems that an existing bonding silver wire is high in hardness, low in tensile strength, prone to being damaged in the bonding process, not firm in bonding and not prone to damage in the bonding process can be solved. And the service life of the bonding silver wire cannot be accurately estimated.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor technology, and in particular to a bonding silver wire life prediction method and system based on big data. Background Art

[0002] Bonding wires are fine metal wire inner leads used to achieve electrical connection between the input / output bonding points of the circuit inside the chip and the internal contact points of the lead frame (or substrate) when assembling semiconductor devices and integrated circuits. The quality of the bonding effect directly affects the performance of the integrated circuit.

[0003] At present, the bonding wires used in the market are mainly divided into gold bonding wire, copper bonding wire and silver bonding wire. Among them, copper wire is mainly used for low- and medium-end circuits, while gold and silver precious metal wires are mainly used for medium- and high-end circuits. Gold bonding wire has good electrical and thermal conductivity and chemical stability, but its cost is relatively high. Compared with gold bonding wire, silver bonding wire (i.e. bonding silver wire) is easy to be damaged during the bonding process due to its high hardness and low tensile strength, and the bonding is not firm, which makes it impossible to accurately estimate the service life of the bonding silver wire. Summary of the invention

[0004] The embodiment of the present application provides a method and system for predicting the life of a silver bonding wire based on big data, which can solve the problem that the existing silver bonding wire (i.e., bonding silver wire) is easily damaged during the bonding process and the bonding is not firm, resulting in the inability to accurately predict the service life of the bonding silver wire due to its high hardness and low tensile strength.

[0005] In a first aspect, an embodiment of the present application provides a bonding silver wire life prediction method based on big data, which is applied to bonding solder joints and bonding silver wires after a bonding process, including: Establishing a solder joint bonding life model based on the data information of the plurality of bonding solder joints; Acquiring a bonding solder joint scanning image of the bonding solder joint by a scanning electron microscope; Analyzing the bonding solder joint scanning image to obtain solder joint bonding data; The solder joint bonding data is input into the solder joint bonding life model to obtain the estimated bonding service life of the bonding solder joint.

[0006] In some embodiments, the step of establishing a solder joint life model based on the data information of the plurality of bonding solder joints includes: Dividing the data information of the plurality of bonding solder joints into a training set and a prediction set; Inputting the data information of the training set into the neural network model to be trained to obtain an initial solder joint life model; Inputting the data information of the prediction set into the initial solder joint life model to obtain an initial prediction result; If the initial prediction result meets the preset requirement, the initial solder joint life model is set as the solder joint combination life model.

[0007] In some embodiments, the solder joint data includes solder joint contour data, solder joint center area data, and solder joint ridge area data of the bonding solder joint; The step of inputting the solder joint bonding data into the solder joint bonding life model to obtain the estimated bonding life of the bonding solder joint is specifically as follows: The solder joint contour data, the solder joint center area data, and the solder joint ridge area data are input into the solder joint life model to obtain the estimated bonding service life of the bonding solder joint.

[0008] In some embodiments, the method for predicting the life of a silver bonding wire based on big data further includes: Scanning the bonding solder joints through a transmission electron microscope to obtain a transmission image of the bonding solder joints; Analyzing the transmission image of the bonding solder joint to obtain solder joint interface data; The solder joint transmission data is input into a solder joint interface life model to obtain an estimated interface service life of the bonding solder joint.

[0009] In some embodiments, the solder joint interface data includes solder joint interface contour data and solder joint interface shape data of the bonding solder joint; The solder joint transmission data is input into the solder joint interface life model to obtain the estimated interface life of the bonding solder joint, specifically: The solder joint interface contour data and the solder joint interface shape data are input into a solder joint interface life model to obtain an estimated interface service life of the bonding solder joint.

[0010] In some embodiments, the method for predicting the life of the silver bonding wire based on big data further includes: Obtaining the post-bonding diameters of the bonding silver wire at multiple positions; The post-bonding service life of the silver bonding wire is evaluated based on the plurality of post-bonding diameters and pre-bonding diameters.

[0011] In some embodiments, the method for predicting the life of the silver bonding wire based on big data further includes: Obtaining the bonding resistance of the bonding silver wire based on the solder joint bonding data and the solder joint interface data; Obtaining the working voltage of the bonding silver wire; An expected operating life of the silver bonding wire is evaluated based on the bonding resistance and the operating voltage.

[0012] In some embodiments, the method for predicting the life of the silver bonding wire based on big data further includes: Establishing a molding material injection effect model when the molding material is injected into the bonding silver wire; Obtaining injection temperature, injection pressure and injection time of the molding compound; The injection temperature, the injection pressure and the injection time are input into a molding material injection damage life model to obtain the injection damage life of the bonding silver wire.

[0013] In some embodiments, the method for predicting the life of the silver bonding wire based on big data further includes: The factory service life of the bonding silver wire is determined based on the expected bonding service life, the expected interface service life, the post-bonding service life, the expected working service life and the injection damage service life.

[0014] In a second aspect, the embodiment of the present application provides a bonding silver wire life prediction system based on big data, which is applied to bonding solder joints and bonding silver wires after a bonding process, including: An establishing module, used for establishing a solder joint bonding life model based on the data information of the plurality of bonding solder joints; A scanning module, used for obtaining a bonding solder joint scanning image of the bonding solder joint through a scanning electron microscope; An analysis module, used for analyzing the bonding solder joint scanning image to obtain solder joint bonding data; The prediction module is used to input the solder joint bonding data into the solder joint bonding life model to obtain the estimated bonding service life of the bonding solder joint.

[0015] Compared with the prior art, the advantages of the embodiments of the present application are that a solder joint bonding life model is established based on the data information of a plurality of the bonding solder joints; a bonding solder joint scanning image of the bonding solder joint is acquired through a scanning electron microscope; the bonding solder joint scanning image is analyzed to obtain solder joint bonding data; the solder joint bonding data is input into the solder joint bonding life model to obtain the estimated bonding service life of the bonding solder joint, so that the service life of the bonding solder joint can be accurately estimated, and the problem that the service life of the existing bonding silver wire cannot be accurately estimated due to its high hardness and low tensile strength, which makes it easy to be damaged during the bonding process and the bonding is not firm. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Hereinafter, the present invention will be described in more detail based on embodiments and with reference to the accompanying drawings.

[0017] Figure 1 It is a flow chart of a bonding silver wire life prediction method based on big data provided by one embodiment of the present invention; Figure 2 It is a schematic diagram of a bonding silver wire life prediction system based on big data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be further described below in conjunction with the accompanying drawings.

[0019] Bonding wire is one of the key materials for electronic packaging, and plays the role of importing and exporting current between the chip and the outside world. At present, the bonding wires used in the market are mainly divided into gold bonding wire, copper bonding wire and silver bonding wire. Among them, copper wire is mainly used for low- and medium-end circuits, while gold and silver precious metal wires are mainly used for medium- and high-end circuits. As a traditional wire bonding material, gold bonding wire has good electrical and thermal conductivity and chemical stability. However, with the rise in gold prices and the continuous development of electronic devices towards miniaturization and multifunctionality, there is an urgent need to reduce packaging costs and increase packaging density. Copper bonding wire and silver bonding wire are cheaper than gold wire and have excellent physical properties, becoming the basic materials to replace gold wire. Compared with gold bonding wire, copper bonding wire and silver bonding wire (i.e. bonding silver wire) still have many disadvantages. For example, copper bonding wire is easy to oxidize, has high hardness, easily causes damage to chips during the bonding process, and is mainly used in low- and medium-end circuits. Silver bonding wire has become the first choice to replace gold bonding wire in medium- and high-end circuits. However, compared with gold bonding wire, silver bonding wire (i.e. bonding silver wire) has high hardness and low tensile strength. It is easy to be damaged during the bonding process and the bonding is not firm, resulting in the inability to accurately estimate the service life of the silver bonding wire.

[0020] First, as Figure 1 As shown, in view of the above technical problems, the embodiment of the present application provides a bonding silver wire life prediction method based on big data, which is applied to bonding solder joints and bonding silver wires after the bonding process, including: S101: Establishing a solder joint bonding life model based on data information of a plurality of bonding solder joints; S102: Acquire a bonding solder joint scanning image of the bonding solder joint through a scanning electron microscope; S103: Analyze the bonding solder joint scanning image to obtain solder joint bonding data; S104: Inputting the solder joint bonding data into the solder joint bonding life model to obtain an estimated bonding life of the bonding solder joint.

[0021] It should be noted that the connection between the silver bonding wire and the pad (i.e., the bonding solder joint) is a micro-connection. In the use of the silver bonding wire, the bonding solder joint is usually more prone to quality problems than the silver bonding wire itself, so it is necessary to analyze the service life of the bonding solder joint in a targeted manner. The silver bonding wire is usually acted on the connection between the silver bonding wire and the pad by ultrasound, electrical energy, thermal energy, etc., so that the silver bonding wire and the pad are combined together. The bonding parameters during the bonding process (such as ultrasonic power, bonding pressure, bonding time) affect the bonding quality of the bonding solder joint, and the bonding quality of the bonding solder joint can be reflected by the bonding situation at the bonding solder joint (i.e., the solder joint bonding data). Therefore, the expected bonding service life of the bonding solder joint can be determined by the solder joint bonding data.

[0022] It should be noted that a scanning electron microscope uses a focused, narrow high-energy electron beam to scan samples (or bonding solder joints), and stimulates various physical information through the interaction between the beam and the material. This information is collected, amplified, and re-imaged to achieve the purpose of characterizing the microscopic morphology of the material (i.e., the scanned image of the bonding solder joint).

[0023] In some embodiments, the step of establishing a solder joint life model based on the data information of the plurality of bonding solder joints includes: Dividing the data information of the plurality of bonding solder joints into a training set and a prediction set; Inputting the data information of the training set into the neural network model to be trained to obtain an initial solder joint life model; Inputting the data information of the prediction set into the initial solder joint life model to obtain an initial prediction result; If the initial prediction result meets the preset requirement, the initial solder joint life model is set as the solder joint combination life model.

[0024] It should be noted that the solder joint life model in the present application is modeled based on big data, and by dividing the data information of the bonding solder joint into a training set and a prediction set, using the training set to train the model, and using the prediction set to test the initial solder joint life model, the accuracy of the solder joint life model can be ensured.

[0025] In some embodiments, the solder joint data includes solder joint contour data, solder joint center area data, and solder joint ridge area data of the bonding solder joint; The step of inputting the solder joint bonding data into the solder joint bonding life model to obtain the estimated bonding life of the bonding solder joint is specifically as follows: The solder joint contour data, the solder joint center area data, and the solder joint ridge area data are input into the solder joint life model to obtain the estimated bonding service life of the bonding solder joint.

[0026] It should be noted that the main characterization morphology of the bonding solder joint is divided into the solder joint bonding contour (or solder joint bonding contour data), such as circle, ellipse, solder joint bonding center area (or solder joint bonding center area data), and solder joint bonding ridge area (or solder joint bonding ridge area data). Usually, the stress in the center area of ​​the bonding solder joint is the most concentrated and plastic flow is not easy to occur. The oxide formed on the bonding wire and the pad surface is difficult to remove, which hinders the bonding of the center area. Therefore, the bonding force provided by the center area of ​​the bonding solder joint is relatively small (or the strength is relatively small); the ridge area of ​​the bonding solder joint provides the main strength of the bonding. degree; the outline of the bonding solder joint usually characterizes the area of ​​the central area of ​​the bonding solder joint, that is, when the outline of the bonding solder joint is circular, the central area of ​​the bonding solder joint is small, and the central area can provide better strength; when the outline of the bonding solder joint is elliptical, the central area of ​​the bonding solder joint is large, and the central area cannot provide better strength. Therefore, the expected bonding service life of the bonding solder joint can be obtained by analyzing the solder joint bonding outline data, the solder joint bonding central area data, and the solder joint bonding ridge area data through the solder joint bonding life model, and combining the loss during use.

[0027] It should be noted that, taking ultrasonic bonding as an example, in terms of ultrasonic power and bonding strength: as the power increases, the plastic deformation of the bonding wire is more complete, and the morphology of the bonding interface gradually changes from an ellipse to a circle, so that the effective connection part gradually expands toward the center, thereby improving the bonding strength. However, when the bonding strength reaches a certain value, if the ultrasonic power continues to increase, the bonding strength will decrease significantly. In addition, excessive ultrasonic power will cause cracks or hardening in the formed bonding interface. In terms of bonding time and bonding strength: if the bonding time is too short, the contaminants on the bonding interface will not be removed in time or the atoms at the interface will not diffuse sufficiently. At the same time, the unbonded area will expand, resulting in a loose bonding surface. As the bonding time increases, the bonding wire As the absorbed energy increases, the lateral deformation rate of the bonding point increases, and the bonding ridge area gradually extends toward the center area, thereby improving the bonding strength. However, if the bonding time is too long, the diameter of the bonding point will be too large, exceeding the edge of the pad, and at the same time, the adhesion between the lead and the splitter will be enhanced, and the bonding will also be weak. In terms of bonding pressure and bonding strength: excessive bonding pressure will hinder the movement of the splitter, resulting in the inability of ultrasonic energy to be transmitted to the bonding interface through the splitter to form a metallurgical bonding surface. At the same time, the contaminants on the interface are squeezed to the bonding center, increasing the unbonded area in the center, and the bonding strength of the interface cannot be guaranteed. Therefore, the bonding parameters affect the bonding quality (or bonding quality) of the bonding solder joint, and thus affect the bonding strength of the bonding solder joint.

[0028] It should be noted that an image analysis sub-model can be trained to perform feature extraction on the bonding solder joint scanning image, so as to more quickly obtain the solder joint joint contour data, the solder joint joint center area data, and the solder joint joint ridge area data. The image analysis sub-model can be a neural network model.

[0029] In some embodiments, the method for predicting the life of a silver bonding wire based on big data further includes: Scanning the bonding solder joints through a transmission electron microscope to obtain a transmission image of the bonding solder joints; Analyzing the transmission image of the bonding solder joint to obtain solder joint interface data; The solder joint transmission data is input into a solder joint interface life model to obtain an estimated interface service life of the bonding solder joint.

[0030] It should be noted that when the bonding silver wire is bonded to the pad, an intermetallic compound will be formed between the bonding silver wire and the pad, and the morphology of the intermetallic compound will also affect the strength of the bonding solder joint. Therefore, by obtaining the solder joint interface data, the influence of the interface morphology of the bonding solder joint on the service life of the bonding solder joint can be estimated, that is, the expected interface service life.

[0031] It should be noted that the transmission electron microscope (TEM) projects an accelerated and focused electron beam onto a very thin sample (bonding solder joint). The electrons collide with atoms in the sample and change direction, thereby producing three-dimensional angular scattering, which can form images of different brightness (i.e., the transmission image of the bonding solder joint). The image will be displayed on an imaging device (such as a fluorescent screen, film, and photosensitive coupling component) after being magnified and focused, so that the interface data of the bonding solder joint can be collected.

[0032] In some embodiments, the solder joint interface data includes solder joint interface contour data and solder joint interface shape data of the bonding solder joint; The solder joint transmission data is input into the solder joint interface life model to obtain the estimated interface life of the bonding solder joint, specifically: The solder joint interface contour data and the solder joint interface shape data are input into a solder joint interface life model to obtain an estimated interface service life of the bonding solder joint.

[0033] It should be noted that the thickness of the intermetallic compound can usually provide better adhesion when it is within a certain thickness range, while the thickness of the intermetallic compound that is too large or too small will reduce the adhesion. The thickness of the intermetallic compound can be obtained through the solder joint interface profile data; usually, the intermetallic compound has better adhesion when it is planar in shape, and has poor adhesion when it is thorny in shape. Therefore, based on the solder joint interface profile data and the solder joint interface shape data, the influence of the intermetallic compound on the service life of the bonding solder joint (i.e., the expected interface service life) can be determined.

[0034] In some embodiments, the method for predicting the life of the silver bonding wire based on big data further includes: Obtaining the post-bonding diameters of the bonding silver wire at multiple positions; The post-bonding service life of the silver bonding wire is evaluated based on the plurality of post-bonding diameters and pre-bonding diameters.

[0035] It should be noted that when the bonding process is carried out, the bonding silver wire will be pulled, which will cause initial damage to the bonding silver wire before it is put into use, thereby affecting the service life of the bonding silver wire. By evaluating the diameter changes at multiple positions on the bonding silver wire, the damage condition of the bonding silver wire can be evaluated, and then the post-bonding service life of the bonding silver wire can be determined in combination with the factory service life of the bonding silver wire (that is, the factory service life of the bonding silver wire minus the life affected by the damage to obtain the post-bonding service life).

[0036] It should be noted that a post-bonding life assessment model can be trained, and the life of the bonding silver wire after bonding can be assessed by the post-bonding assessment model, thereby improving the assessment speed and accuracy.

[0037] In some embodiments, the method for predicting the life of the silver bonding wire based on big data further includes: Obtaining the bonding resistance of the bonding silver wire based on the solder joint bonding data and the solder joint interface data; Obtaining the working voltage of the bonding silver wire; An expected operating life of the silver bonding wire is evaluated based on the bonding resistance and the operating voltage.

[0038] It should be noted that the bonding condition and interface condition of the bonding solder joint will affect the effective contact area between the bonding silver wire and the pad in the bonding solder joint, and the change in the effective contact area will affect the resistance of the bonding solder joint. Usually, the service life of the bonding solder joint is the service life under standard working conditions, and the impact of the resistance change of the bonding solder joint on its service life is not taken into account. The embodiment of the present application can determine the bonding condition and interface condition of the bonding solder joint through the solder joint bonding data and the solder joint interface data, thereby making an accurate estimate of the resistance of the bonding solder joint, and then combined with the working voltage of the bonding silver wire, it can determine the expected working life of the bonding solder joint.

[0039] It should be noted that by analyzing the change in diameter of the bonding silver wire after bonding, the actual resistance of the bonding silver wire can be determined, and the expected working life of the bonding silver wire can be determined based on the actual resistance of the bonding silver wire and the working voltage of the bonding silver wire.

[0040] In some embodiments, the method for predicting the life of the silver bonding wire based on big data further includes: Establishing a molding material injection effect model when the molding material is injected into the bonding silver wire; Obtaining injection temperature, injection pressure and injection time of the molding compound; The injection temperature, the injection pressure and the injection time are input into a molding material injection damage life model to obtain the injection damage life of the bonding silver wire.

[0041] It should be noted that, usually when the chip is subjected to bonding process and packaging, it is necessary to inject molding compound into the chip, and when injecting the molding compound, it is necessary to apply a certain pressure (i.e., the injection pressure), temperature (i.e., the injection temperature), and an action time (i.e., the injection time) to the molding compound, and the injection of molding compound will have an impact on the strength of the bonding solder joints and the bonding silver wires, thereby affecting the life of the bonding solder joints and the bonding silver wires. Usually, the greater the injection pressure, the higher the injection temperature, and the longer the injection time, the greater the impact on the life of the bonding solder joints and the bonding silver wires. Therefore, by establishing the molding compound injection impact model, the injection damage life of the bonding silver wire can be accurately estimated, wherein the molding compound injection impact model can be a trained neural network model.

[0042] In some embodiments, the method for predicting the life of the silver bonding wire based on big data further includes: The factory service life of the bonding silver wire is determined based on the expected bonding service life, the expected interface service life, the post-bonding service life, the expected working service life and the injection damage service life.

[0043] It should be noted that when determining the factory service life of the bonding silver wire, the minimum value is usually taken among the expected bonding service life, the expected interface service life, the post-bonding service life, and the expected working life, and then the injection damage life is subtracted to obtain the factory service life.

[0044] It should be noted that a molding material aging impact model can be established to analyze the impact of the molding material aging on the life of the bonding silver wire and the bonding solder joint, that is, the aging damage life, and then the aging damage life is pre-added when calculating the factory service life, thereby accurately estimating the factory service life of the bonding silver wire.

[0045] To sum up, the advantages of the embodiments of the present application are that a solder joint bonding life model is established based on the data information of multiple bonding solder joints; a bonding solder joint scanning image of the bonding solder joint is obtained through a scanning electron microscope; the bonding solder joint scanning image is analyzed to obtain solder joint bonding data; the solder joint bonding data is input into the solder joint bonding life model to obtain the expected bonding service life of the bonding solder joint, so that the service life of the bonding solder joint can be accurately estimated, and the problem that the existing bonding silver wire cannot be accurately estimated due to its high hardness and low tensile strength, which makes it easy to be damaged during the bonding process and the bonding is not firm.

[0046] Second, as Figure 2 As shown, the embodiment of the present application provides a bonding silver wire life prediction system based on big data, which is applied to bonding solder joints and bonding silver wires after a bonding process, including: An establishing module 210 is used to establish a solder joint bonding life model based on the data information of the plurality of bonding solder joints; A scanning module 220, configured to obtain a bonding solder joint scanning image of the bonding solder joint through a scanning electron microscope; An analysis module 230 is used to analyze the bonding solder joint scanning image to obtain solder joint bonding data; The prediction module 240 is used to input the solder joint bonding data into the solder joint bonding life model to obtain the estimated bonding service life of the bonding solder joint.

[0047] In some embodiments, the step of establishing a solder joint life model based on the data information of the plurality of bonding solder joints includes: Dividing the data information of the plurality of bonding solder joints into a training set and a prediction set; Inputting the data information of the training set into the neural network model to be trained to obtain an initial solder joint life model; Inputting the data information of the prediction set into the initial solder joint life model to obtain an initial prediction result; If the initial prediction result meets the preset requirement, the initial solder joint life model is set as the solder joint combination life model.

[0048] In some embodiments, the solder joint data includes solder joint contour data, solder joint center area data, and solder joint ridge area data of the bonding solder joint; The step of inputting the solder joint bonding data into the solder joint bonding life model to obtain the estimated bonding life of the bonding solder joint is specifically as follows: The solder joint contour data, the solder joint center area data, and the solder joint ridge area data are input into the solder joint life model to obtain the estimated bonding service life of the bonding solder joint.

[0049] In some embodiments, the prediction module 240 is further configured to: Scanning the bonding solder joints through a transmission electron microscope to obtain a transmission image of the bonding solder joints; Analyzing the transmission image of the bonding solder joint to obtain solder joint interface data; The solder joint transmission data is input into a solder joint interface life model to obtain an estimated interface service life of the bonding solder joint.

[0050] In some embodiments, the solder joint interface data includes solder joint interface contour data and solder joint interface shape data of the bonding solder joint; The solder joint transmission data is input into the solder joint interface life model to obtain the estimated interface life of the bonding solder joint, specifically: The solder joint interface contour data and the solder joint interface shape data are input into a solder joint interface life model to obtain an estimated interface service life of the bonding solder joint.

[0051] In some embodiments, the prediction module 240 is further configured to: Obtaining the post-bonding diameters of the bonding silver wire at multiple positions; The post-bonding service life of the silver bonding wire is evaluated based on the plurality of post-bonding diameters and pre-bonding diameters.

[0052] In some embodiments, the prediction module 240 is further configured to: Obtaining the bonding resistance of the bonding silver wire based on the solder joint bonding data and the solder joint interface data; Obtaining the working voltage of the bonding silver wire; An expected operating life of the silver bonding wire is evaluated based on the bonding resistance and the operating voltage.

[0053] In some embodiments, the prediction module 240 is further configured to: Establishing a molding material injection effect model when the molding material is injected into the bonding silver wire; Obtaining injection temperature, injection pressure and injection time of the molding compound; The injection temperature, the injection pressure and the injection time are input into a molding material injection damage life model to obtain the injection damage life of the bonding silver wire.

[0054] In some embodiments, the prediction module 240 is further configured to: The factory service life of the bonding silver wire is determined based on the expected bonding service life, the expected interface service life, the post-bonding service life, the expected working service life and the injection damage service life.

[0055] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0056] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0057] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0058] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for predicting the life of silver bonding wire based on big data, characterized in that: Applied to bonding solder joints and bonding silver wires after bonding process, including: Establishing a solder joint bonding life model based on the data information of the plurality of bonding solder joints; Acquiring a bonding solder joint scanning image of the bonding solder joint by a scanning electron microscope; Analyzing the bonding solder joint scanning image to obtain solder joint bonding data; The solder joint bonding data is input into the solder joint bonding life model to obtain the estimated bonding service life of the bonding solder joint.

2. The method for predicting the life of a silver bonding wire based on big data according to claim 1, characterized in that: The step of establishing a solder joint bonding life model based on the data information of the plurality of bonding solder joints comprises: Dividing the data information of the plurality of bonding solder joints into a training set and a prediction set; Inputting the data information of the training set into the neural network model to be trained to obtain an initial solder joint life model; Inputting the data information of the prediction set into the initial solder joint life model to obtain an initial prediction result; If the initial prediction result meets the preset requirement, the initial solder joint life model is set as the solder joint combination life model.

3. The method for predicting the life of a silver bonding wire based on big data according to claim 1, characterized in that: The solder joint data includes solder joint contour data, solder joint center area data, and solder joint ridge area data of the bonding solder joint; The step of inputting the solder joint bonding data into the solder joint bonding life model to obtain the estimated bonding life of the bonding solder joint is specifically as follows: The solder joint contour data, the solder joint center area data, and the solder joint ridge area data are input into the solder joint life model to obtain the estimated bonding service life of the bonding solder joint.

4. The method for predicting the life of a silver bonding wire based on big data according to claim 1, characterized in that: Also includes: Scanning the bonding solder joints through a transmission electron microscope to obtain a transmission image of the bonding solder joints; Analyzing the transmission image of the bonding solder joint to obtain solder joint interface data; The solder joint transmission data is input into a solder joint interface life model to obtain an estimated interface service life of the bonding solder joint.

5. The bonding silver wire life prediction method based on big data according to claim 4 is characterized in that: The solder joint interface data includes solder joint interface contour data and solder joint interface shape data of the bonding solder joint; The solder joint transmission data is input into the solder joint interface life model to obtain the estimated interface life of the bonding solder joint, specifically: The solder joint interface contour data and the solder joint interface shape data are input into a solder joint interface life model to obtain an estimated interface service life of the bonding solder joint.

6. The bonding silver wire life prediction method based on big data according to claim 4 is characterized in that: Also includes: Obtaining the post-bonding diameters of the bonding silver wire at multiple positions; The post-bonding service life of the silver bonding wire is evaluated based on the plurality of post-bonding diameters and pre-bonding diameters.

7. The method for predicting the life of a silver bonding wire based on big data according to claim 6, characterized in that: Also includes: Obtaining the bonding resistance of the bonding silver wire based on the solder joint bonding data and the solder joint interface data; Obtaining the working voltage of the bonding silver wire; An expected operating life of the silver bonding wire is evaluated based on the bonding resistance and the operating voltage.

8. The method for predicting the life of silver bonding wire based on big data according to claim 7, characterized in that: Also includes: Establishing a molding material injection effect model when the molding material is injected into the bonding silver wire; Obtaining injection temperature, injection pressure and injection time of the molding compound; The injection temperature, the injection pressure and the injection time are input into a molding material injection damage life model to obtain the injection damage life of the bonding silver wire.

9. The method for predicting the life of a silver bonding wire based on big data according to claim 8, characterized in that: Also includes: The factory service life of the bonding silver wire is determined based on the expected bonding service life, the expected interface service life, the post-bonding service life, the expected working service life and the injection damage service life.

10. A bonding silver wire life prediction system based on big data, characterized in that: Applied to bonding solder joints and bonding silver wires after bonding process, including: An establishing module, used for establishing a solder joint bonding life model based on the data information of the plurality of bonding solder joints; A scanning module, used for obtaining a bonding solder joint scanning image of the bonding solder joint through a scanning electron microscope; An analysis module, used for analyzing the bonding solder joint scanning image to obtain solder joint bonding data; The prediction module is used to input the solder joint bonding data into the solder joint bonding life model to obtain the estimated bonding service life of the bonding solder joint.

Images