Welding device for chip production and packaging and control method thereof

Abstract

The invention relates to the technical field of chip production equipment, in particular to a welding device for chip production packaging and a control method of the welding device. The interference area between the abnormal moisture migration area and the maximum susceptibility range is obtained, the welding loss probability that the moisture invasion sensitive chip is damaged due to the abnormal moisture migration area is calculated based on the interference area and the vulnerability index of the moisture sensitive chip, and the chip conveying mechanism is controlled to preheat the moisture invasion sensitive chip according to the welding loss probability; and distributing a steady-state control voltage of each circuit component in the accurate circuit control topological graph based on the expected control logic effect to obtain a steady-state drift difference value of each chip control node, and according to the steady-state drift difference value, deciding a welding process of a packaging type and planning welding scheduling of subordinate welding equipment. According to the invention, precise element preheating control can be carried out on the chip affected by air humidity in the chip packaging and welding process, and the accuracy, efficiency and qualified rate of chip packaging and welding are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of chip production equipment, and in particular to a welding device for chip production packaging and a control method thereof. Background Art

[0002] Welding is the connection process between chip and package, and welding control in chip production and packaging is an important link to ensure the quality of packaging and improve product reliability. Commonly used welding methods include lead-tin welding, lead-free welding, ball welding (BGA), etc. Through refined process control and advanced detection technology to perform the above welding methods, the welding quality of chip packaging can be effectively improved to ensure the reliability and stability of the product. However, the existing chip production and packaging welding equipment has low welding control accuracy, and there is still a large packaging welding error during the welding process. In addition, when the existing equipment transports the chip to the welding table for packaging, it is impossible to detect and judge whether the surface of the chip is attached to the moisture in the surrounding humidity air, making it difficult to achieve the effect of preheating the chip components. It is easy to cause water vapor explosion during welding, resulting in chip damage, which greatly increases the scrap rate of chip packaging welding. In addition, the existing equipment has low scheduling efficiency for welding guns and other mechanisms, and the welding time cost is high, which is not conducive to meeting the production welding needs of high-quality chip packaging. Summary of the invention

[0003] The present invention overcomes the shortcomings of the prior art and provides a chip production packaging welding device and a control method thereof.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] A first aspect of the present invention provides a welding control method for chip production packaging, comprising the following steps:

[0006] S102: Obtain the susceptible point layout of the moisture-sensitive chip in the chip batch transportation area and the detection range and distribution pattern diagram of the humidity sensor, constrain the moisture-sensitive chip to be in the maximum susceptible range in the chip batch transportation area based on the susceptible point layout, detection range and distribution pattern diagram, and perform dynamic transportation calculation to obtain a dynamic transportation model in which the maximum susceptible range of the moisture-sensitive chip is in the chip batch transportation area;

[0007] S104: Acquire several groups of humidity data of the humidity sensor in the virtual transportation space, interpolate each group of humidity data into the Thiessen polygon of the virtual transportation space for migration estimation, so as to describe the moisture migration simulation field around the layout of susceptible points in the chip batch transportation area, and combine the moisture migration simulation field with the dynamic transportation model interference analysis to obtain the chip humidity detection result;

[0008] S106: Obtaining the interference area between the abnormal moisture migration area and the maximum susceptible range, calculating the soldering probability of the moisture invasion sensitive chip being damaged due to the abnormal moisture migration area based on the interference area and the vulnerability index of the moisture invasion sensitive chip, and controlling the chip transport mechanism to preheat the moisture invasion sensitive chip according to the soldering probability;

[0009] S108: Construct an accurate circuit control topology diagram in which the designated welding area of ​​a given chip is located on the PCB circuit board, allocate the steady-state control voltage of each circuit component in the accurate circuit control topology diagram based on the expected control logic effect, obtain the steady-state drift difference of each chip control node, and decide the welding process of the package type according to the steady-state drift difference and plan the welding scheduling of the subordinate welding equipment.

[0010] More specifically, the step S102 includes the following steps:

[0011] Obtain the predetermined welding requirements for chip production and packaging and the chip batch transportation area, and extract several predetermined chips that must be welded on the target PCB substrate based on the predetermined welding requirements;

[0012] Extracting and comparing the single pin features of all the given chips transported in the chip batch transport area, if the similarity between the single pin feature and any reference pin feature in the standard pin comparison warehouse is greater than a preset similarity, the given chip is marked as a moisture sensitive chip;

[0013] Acquire architecture information of the moisture-sensitive chip, search in a big data network based on the architecture information to identify susceptible points and susceptible point layout of the moisture-sensitive chip, and preset a migration interference range threshold based on the susceptible point layout;

[0014] Construct a virtual transport space domain for the chip batch transport area, obtain the detection range of the humidity sensor and the distribution pattern diagram of the humidity sensor in the virtual transport space domain, and set the space humidity perception coverage limit according to the distribution pattern diagram;

[0015] Taking the susceptible range of the humidity sensitive chip in the virtual transportation space as the termination function, constraining the termination function based on the migration interference range threshold until the constrained termination function approaches the perception coverage blind area covered by the space humidity perception coverage boundary, and obtaining the maximum susceptible range;

[0016] The image data of several frames of humidity-sensitive chips transported in the chip batch transportation area are obtained by shooting with industrial cameras with preset uniform timestamps, and the preset uniform timestamps are decomposed by STL to decompose them into three parts: trend, seasonality and residual.

[0017] Based on the image data on several frames, a model of the trend part of the transportation of humidity-sensitive chips in the chip batch transportation area is built to capture the dynamic pattern of the residual part, and a trend-residual initial model is obtained. The timestamp time series of the seasonal part is added to the trend-residual initial model, and finally a dynamic transportation model is generated in which the maximum susceptible range of humidity-sensitive chips is in the chip batch transportation area.

[0018] More specifically, the step S104 includes the following steps:

[0019] Migration sampling points are set based on the layout of susceptible points, and several groups of humidity data detected by humidity sensors at migration sampling points in the virtual transportation space at preset uniform timestamps are obtained through the Internet of Things;

[0020] The virtual transportation space domain is planned into N sub-domains according to the migration sampling points, a Thiessen polygon is constructed based on each sub-domain, an interpolation point of each humidity data is defined, and the interpolation points of several groups of humidity data detected by each migration sampling point are interpolated on the corresponding Thiessen polygon to be interpolated points;

[0021] After the interpolation is completed, a new Thiessen polygon will be generated for each Thiessen polygon. The overlapping area between the new Thiessen polygon after the interpolation point is inserted and the surrounding original Thiessen polygons is obtained. The ratio of the corresponding overlapping area of ​​the interpolation point and the adjacent interpolation point to the total area of ​​the new Thiessen polygon is calculated to obtain the overlapping area weight value of each interpolation point.

[0022] The weighted average of the humidity data of each interpolation point is calculated according to the overlapping area weight value, and the migration estimation value of all interpolation points is output. The moisture migration simulation field around the layout of susceptible points in the chip batch transportation area is constructed by describing the migration estimation value of all interpolation points;

[0023] Based on the moisture migration simulation field, a moisture migration simulation area corresponding to an absolute humidity greater than a preset absolute humidity is stripped off, which is defined as an abnormal moisture migration area, and a dynamic transport area of ​​the maximum susceptible range of the moisture sensitive chip is obtained through the dynamic transport model;

[0024] If there is no regional interference between the abnormal moisture migration area and the dynamic transport area, the humidity sensitive chip is calibrated as a healthy humidity sensitive chip; if there is regional interference between the abnormal moisture migration area and the dynamic transport area, the humidity sensitive chip is calibrated as a moisture invasion sensitive chip, and the chip humidity detection result is obtained.

[0025] More specifically, the step S106 includes the following steps:

[0026] The chip transport mechanism detects whether moisture intrusion occurs in the chip batch transport area and the sensitive chip. If so, the control element preheating mechanism is started, and the interference area between the abnormal moisture migration area and the dynamic transport area of ​​the maximum susceptible range of the moisture sensitive chip is obtained;

[0027] Based on the absolute humidity value, the environmental index of the water vapor content in the abnormal moisture migration area is calculated in the moisture migration simulation field to obtain the water vapor content density in the abnormal moisture migration area;

[0028] Based on the architecture information of the moisture-sensitive chip, the circuit structure of the moisture-sensitive chip is drawn, and a chip architecture knowledge graph is obtained, the circuit structure and the water vapor content density are identified in the chip architecture knowledge graph, and a vulnerability index of welding of the moisture-sensitive chip under the water vapor content density condition is obtained;

[0029] Preset an integral interval according to the vulnerability index, and generate multiple random sampling points by randomly extracting a probability density function within the interference area to ensure that the integral of the randomly extracted probability density function is within the integral interval;

[0030] Calculate the average value of the probability density function of each random sampling point, and add the average values ​​of all random sampling points to obtain an integral approximation, and determine the probability of soldering damage in the soldering process of the moisture-sensitive chip caused by the water vapor content density in the abnormal moisture migration area according to the integral approximation;

[0031] If the solder loss probability is greater than the preset solder loss probability, the chip transport mechanism is controlled to transfer the moisture-sensitive chip to the component preheating mechanism, and the preheating temperature of the component preheating mechanism is controlled based on the solder loss probability to evaporate and remove moisture on the surface of the chip affected by moisture.

[0032] More specifically, the step S108 includes the following steps:

[0033] Obtain a designated soldering area on a PCB circuit board reserved for soldering a predetermined chip, and obtain a circuit control architecture and an expected control logic effect of the designated soldering area on the PCB circuit board;

[0034] Kirchhoff's law is introduced, and an accurate circuit control topology map of the specified welding area is constructed in Kirchhoff's law based on the circuit control architecture, and the chip control nodes are extracted through the accurate circuit control topology map;

[0035] Obtain a global circuit control topology of a PCB circuit board, divide the global circuit control topology into a plurality of sub-topology regions, obtain model information of each circuit component, and perform a relaxation planning of a steady-state control voltage integer solution of each circuit component in each sub-topology region using the expected control logic effect and the model information as constraint criteria to obtain an integer programming model;

[0036] Preset an integer solution for a non-steady-state control voltage, obtain a cutting plane of a non-integer constraint contained in an integer programming model, add the cutting plane to a constraint set of the integer programming model and solve it to obtain a new integer solution for a steady-state control voltage;

[0037] If the new steady-state control voltage integer solution still contains a non-integer solution threshold, the cutting plane adding and solving steps are repeated until the non-integer solution threshold is contained, so as to iteratively approximate the steady-state control voltage integer solution of the expected control logic effect, obtain the sub-steady-state control voltage of each circuit component in each sub-topology area, and calculate the expected steady-state voltage value corresponding to each chip control node by combining the sub-steady-state control voltages of all circuit components;

[0038] Creating a control simulation model of the circuit control architecture, using the control simulation model to simulate the circuit control after a given chip is welded in a designated welding area, generating a circuit simulation control result, and obtaining an actual steady-state voltage value of each chip control node through the circuit simulation control result;

[0039] Obtaining a package type of a given chip, calculating a steady-state drift difference between the expected steady-state voltage value and the actual steady-state voltage value, formulating and improving a welding process for the package type according to the steady-state drift difference, and generating a package welding process plan;

[0040] The pin shape, pin number and pin spacing of a given chip are obtained, and the subordinate welding equipment is obtained according to the pin shape and the package welding process plan. The welding schedule of each subordinate welding equipment for each welding point is planned based on the pin spacing to obtain the welding scheduling network.

[0041] More specifically, in the step S108, the pin shape, pin number and pin spacing of a given chip are obtained, the subordinate welding equipment is obtained according to the pin shape and the package welding process plan, and the welding schedule of each subordinate welding equipment for each welding point is planned based on the pin spacing to obtain a welding scheduling network, which specifically includes the following steps:

[0042] Obtain the pin shape, pin number and pin spacing of a given chip, and also obtain the layout design information of the given chip in the specified soldering area on the PCB circuit board;

[0043] Determine the welding equipment required for each welding point on a given chip according to the packaging welding process scheme and the pin shape, define it as subordinate welding equipment, and obtain the historical welding conditions of each subordinate welding equipment for the given chip;

[0044] The historical welding conditions of each subordinate welding equipment are analyzed based on the number of pins to obtain the welding frequency of each subordinate welding equipment. The singular value decomposition algorithm is introduced to perform singular value decomposition on the matrix corresponding to the welding frequency of each subordinate welding equipment to obtain the singular value matrix of the subordinate welding equipment.

[0045] Based on the singular value matrix of the subordinate welding equipment, M singular values ​​are selected to reconstruct the matrix corresponding to the welding frequency of each subordinate welding equipment to determine the singular weight value of each subordinate welding equipment;

[0046] A frequent singular weight threshold is preset. If the singular weight value does not exceed the frequent singular weight threshold, the pin welding points corresponding to one or more subordinate welding devices of the singular weight value are marked as idle welding points; if the singular weight value exceeds the frequent singular weight threshold, the pin welding points corresponding to one or more subordinate welding devices of the singular weight value are marked as frequent welding points;

[0047] According to the singular weight values ​​of idle welding points and frequent welding points, an idle-frequent welding point layout diagram is fitted in the layout design information and the pin spacing. According to the idle-frequent welding point layout diagram, the welding scheduling network of each subordinate welding equipment is planned, and the welding scheduling network is uploaded to the chip packaging welding control terminal.

[0048] A second aspect of the present invention provides a chip production and packaging welding device, which is applied to any one of the chip production and packaging welding control methods, and specifically includes:

[0049] A chip loading tray, wherein the chip loading tray is used to transport predetermined chips in batches;

[0050] A chip transportation monitoring module, wherein the chip transportation monitoring module includes an industrial camera, which is responsible for capturing image data of a plurality of frames in a time sequence of humidity-sensitive chips transported in a chip batch transportation area;

[0051] A humidity detection module, the humidity detection module includes a humidity sensor and an Internet of Things, and is used to obtain a plurality of groups of humidity data of a virtual transport space migration sampling point at a preset uniform time stamp of a moisture sensitive chip;

[0052] A component preheating module, the component preheating module is used to preheat the moisture-sensitive chip to remove water vapor components on its surface;

[0053] A moisture migration simulation field construction module, wherein the moisture migration simulation field construction module is used to construct a moisture migration simulation field around the susceptible point layout in the chip batch transportation area according to the interpolation migration estimate of the humidity data;

[0054] The scheduling network planning module is responsible for planning the welding scheduling network of each subordinate welding equipment according to the idle-frequent welding point layout diagram.

[0055] The present invention solves the technical defects existing in the background technology, and the beneficial technical effects of the present invention are:

[0056] Based on the layout of susceptible points, detection range and distribution pattern of moisture-sensitive chips in the chip batch transportation area, the maximum susceptible range is constrained and dynamic transportation calculation is performed to obtain a dynamic transportation model in which the maximum susceptible range of moisture-sensitive chips is in the chip batch transportation area; obtain several groups of humidity data of humidity sensors in the virtual transportation space field, build a moisture migration simulation field through the interpolation and migration estimation description of the virtual transportation space field of each group of humidity data, and combine the moisture migration simulation field with the dynamic transportation model for interference analysis to obtain the chip humidity detection result; obtain the interference area between the abnormal moisture migration area and the maximum susceptible range, calculate the soldering probability of moisture invasion sensitive chips caused by abnormal moisture migration area based on the interference area and the vulnerability index of moisture-sensitive chips, and control the chip transportation mechanism to preheat the moisture invasion sensitive chips according to the soldering probability; construct an accurate circuit control topology diagram in which the designated welding area of ​​a given chip welding is located on the PCB circuit board, allocate the steady-state control voltage of each circuit component in the accurate circuit control topology diagram based on the expected control logic effect, obtain the steady-state drift difference of each chip control node, decide the welding process of the packaging type according to the steady-state drift difference, and plan the welding scheduling of the affiliated welding equipment. The present invention can accurately control the preheating of chips affected by air humidity during chip packaging and welding, and simultaneously realize autonomous decision-making of chip packaging technology and efficient planning of welding equipment scheduling, thereby improving the accuracy and efficiency of chip packaging and welding, and ensuring the chip packaging and welding qualification rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0058] Figure 1 A first method flow chart of a welding control method for chip production packaging is shown;

[0059] Figure 2 A second method flow chart of a welding control method for chip production packaging is shown;

[0060] Figure 3The overall structure diagram of a chip production and packaging welding device is shown;

[0061] Figure 4 A partial device structure diagram of a welding device for chip production packaging is shown. DETAILED DESCRIPTION

[0062] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0063] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0064] The first aspect of the present invention provides a welding control method for chip production packaging, such as Figure 1 As shown, the following steps are included:

[0065] S102: Obtain the susceptible point layout of the moisture-sensitive chip in the chip batch transportation area and the detection range and distribution pattern diagram of the humidity sensor, constrain the moisture-sensitive chip to be in the maximum susceptible range in the chip batch transportation area based on the susceptible point layout, detection range and distribution pattern diagram, and perform dynamic transportation calculation to obtain a dynamic transportation model in which the maximum susceptible range of the moisture-sensitive chip is in the chip batch transportation area;

[0066] S104: Acquire several groups of humidity data of the humidity sensor in the virtual transportation space, interpolate each group of humidity data into the Thiessen polygon of the virtual transportation space for migration estimation, so as to describe the moisture migration simulation field around the layout of susceptible points in the chip batch transportation area, and combine the moisture migration simulation field with the dynamic transportation model interference analysis to obtain the chip humidity detection result;

[0067] S106: Obtaining the interference area between the abnormal moisture migration area and the maximum susceptible range, calculating the soldering probability of the moisture invasion sensitive chip being damaged due to the abnormal moisture migration area based on the interference area and the vulnerability index of the moisture invasion sensitive chip, and controlling the chip transport mechanism to preheat the moisture invasion sensitive chip according to the soldering probability;

[0068] S108: Construct an accurate circuit control topology diagram in which the designated welding area of ​​a given chip is located on the PCB circuit board, allocate the steady-state control voltage of each circuit component in the accurate circuit control topology diagram based on the expected control logic effect, obtain the steady-state drift difference of each chip control node, and decide the welding process of the package type according to the steady-state drift difference and plan the welding scheduling of the subordinate welding equipment.

[0069] More specifically, the step S102 includes the following steps:

[0070] Obtain the predetermined welding requirements for chip production and packaging and the chip batch transportation area, and extract several predetermined chips that must be welded on the target PCB substrate based on the predetermined welding requirements;

[0071] Extracting and comparing the single pin features of all the given chips transported in the chip batch transport area, if the similarity between the single pin feature and any reference pin feature in the standard pin comparison warehouse is greater than a preset similarity, the given chip is marked as a moisture sensitive chip;

[0072] Acquire architecture information of the moisture-sensitive chip, search in a big data network based on the architecture information to identify susceptible points and susceptible point layout of the moisture-sensitive chip, and preset a migration interference range threshold based on the susceptible point layout;

[0073] Construct a virtual transport space domain for the chip batch transport area, obtain the detection range of the humidity sensor and the distribution pattern diagram of the humidity sensor in the virtual transport space domain, and set the space humidity perception coverage limit according to the distribution pattern diagram;

[0074] Taking the susceptible range of the humidity sensitive chip in the virtual transportation space as the termination function, constraining the termination function based on the migration interference range threshold until the constrained termination function approaches the perception coverage blind area covered by the space humidity perception coverage boundary, and obtaining the maximum susceptible range;

[0075] The image data of several frames of humidity-sensitive chips transported in the chip batch transportation area are obtained by shooting with industrial cameras with preset uniform timestamps, and the preset uniform timestamps are decomposed by STL to decompose them into three parts: trend, seasonality and residual.

[0076] Based on the image data on several frames, a model of the trend part of the transportation of humidity-sensitive chips in the chip batch transportation area is built to capture the dynamic pattern of the residual part, and a trend-residual initial model is obtained. The timestamp time series of the seasonal part is added to the trend-residual initial model, and finally a dynamic transportation model is generated in which the maximum susceptible range of humidity-sensitive chips is in the chip batch transportation area.

[0077] It should be noted that under certain usage requirements, the packaging of some chips has stricter requirements on their welding environment. Certain environmental factors may cause damage to the chip during the packaging and welding process, especially when using moisture-sensitive components, such as some BGA chips. If there is moisture on the surface of the chip or PCB, the chip package may produce water vapor explosion during welding, causing damage to the chip, greatly increasing the scrap rate of chip packaging welding. Therefore, it is necessary to trace the chips that have surface water vapor components caused by air humidity during chip packaging and transportation. Therefore, this method first screens the chips transported in the chip batch transportation area through the system preset standard pin comparison warehouse to determine the chips that are easily affected by moisture, that is, moisture sensitive chips; since moisture sensitive chips are extremely susceptible to damage due to moisture adhesion, they have multiple moisture susceptible points, and the transportation of chips is a dynamic process, and the influence of moisture in the air on them is also a dynamic behavior. Therefore, the humidity sensor can detect the layout of susceptible points as the target benchmark to constrain the maximum susceptible range of humidity sensitive chips in the chip batch transportation area, that is, based on the migration interference range threshold, the susceptible range is constrained until it approaches the perception coverage blind area covered in the spatial humidity perception coverage boundary, where the perception coverage blind area is beyond the maximum detection range of the humidity sensor in the distribution pattern diagram state, ensuring that the traceability of moisture impact on the chip is always within the humidity monitoring range, ensuring the reliability of the traceability of moisture interference on the chip. Finally, the trend, seasonality and residual calculation of STL decomposition are used to construct a dynamic transportation model in which the maximum susceptible range of humidity sensitive chips is in the chip batch transportation area. This method can construct an accurate and visual dynamic model of the chip transported in a humid environment, providing a reliable analysis basis for whether the chip is affected by moisture during subsequent transportation, and improving the accuracy of tracing the moisture-invaded chip. Among them, the STL decomposition method is a non-parametric time series data decomposition method based on local weighted regression, which is used to decompose time series data into three main components: trend, seasonality, and residual. The STL decomposition method can more clearly understand the structure of the time series and ensure the accuracy of the dynamic transportation expression of the chip in the surrounding environment.

[0078] More specifically, the step S104 is as follows: Figure 2 As shown, the specific steps include:

[0079] S202: setting migration sampling points based on the layout of susceptible points, and obtaining several groups of humidity data detected by the migration sampling points of humidity sensors in the virtual transportation space at preset uniform timestamps through the Internet of Things;

[0080] S204: planning the virtual transport space domain into N sub-domains according to the migration sampling points, constructing a Thiessen polygon based on each sub-domain, defining an interpolation point for each humidity data, and interpolating the interpolation points of several groups of humidity data detected by each migration sampling point on the corresponding Thiessen polygon to be interpolated points;

[0081] S206: After the interpolation is completed, a new Thiessen polygon is generated for each Thiessen polygon, and the overlapping area between the new Thiessen polygon after the interpolation point is inserted and the surrounding original Thiessen polygons is obtained, and the ratio of the corresponding overlapping area of ​​the interpolation point and the adjacent interpolation point to the total area of ​​the new Thiessen polygon is calculated to obtain the overlapping area weight value of each interpolation point;

[0082] S208: Calculate the weighted average of the humidity data of each interpolation point according to the overlapping area weight value, output the migration estimation value of all interpolation points, and build a moisture migration simulation field around the layout of susceptible points in the chip batch transportation area through the migration estimation value description of all interpolation points;

[0083] S210: Based on the moisture migration simulation field, a moisture migration simulation area corresponding to an absolute humidity greater than a preset absolute humidity is stripped off, which is defined as an abnormal moisture migration area, and a dynamic transport area of ​​a maximum susceptible range of a moisture sensitive chip is obtained through the dynamic transport model;

[0084] S212: If there is no regional interference between the abnormal moisture migration area and the dynamic transport area, the humidity sensitive chip is calibrated as a healthy humidity sensitive chip; if there is regional interference between the abnormal moisture migration area and the dynamic transport area, the humidity sensitive chip is calibrated as a moisture invasion sensitive chip to obtain the chip humidity detection result.

[0085] It should be noted that environmental humidity is also a relatively dynamic moment-to-moment change behavior. If the humidity data of the prior art is directly or indirectly fitted and analyzed in the dynamic transportation model, there will be a large model analysis error, resulting in inaccurate chip humidity detection results. Therefore, this method sets interpolation points according to the humidity data detected by the migration sampling points of the humidity sensor in the virtual transportation space field, and converts the virtual transportation space field into Thiessen polygons, and interpolates its interpolation points one by one in the corresponding Thiessen polygons. At this time, the ratio of the overlapping area corresponding to the interpolation point and the adjacent interpolation point to the total area of ​​the new Thiessen polygon is a spatial migration weight value of the humidity data, that is, the overlapping area weight value; through the overlapping area weight value, it is possible to describe the estimated migration change of the humidity data of each interpolation point, thereby building a more accurate moisture migration simulation field around the layout of susceptible points in the chip batch transportation area, and improving the credibility of the subsequent regional interference analysis results. Absolute humidity is an important indicator to measure the content of water vapor in the air. If the absolute humidity is greater than the preset absolute humidity, it means that the water vapor content in the area is high, which is abnormal humidity. If there is no regional interference between the abnormal moisture migration area and the dynamic transportation area, it means that the humidity sensitive chip in the current chip batch transportation area will not be affected by the air humidity, so it is calibrated as a healthy humidity sensitive chip. If there is regional interference, it means that the spatial migration of humidity and water vapor in the air has invaded the humidity sensitive chip, which may cause water vapor components to adhere to its surface, so it is calibrated as a moisture invasion sensitive chip. This method can construct a moisture migration simulation field in a chip transportation environment to perform regional interference analysis on the dynamic transportation model of the chip, so as to accurately determine whether the humidity and water vapor in the air affect the chip in transportation, achieve efficient and accurate rapid detection of chip surface humidity, and improve the control accuracy of chip preheating.

[0086] More specifically, the step S106 includes the following steps:

[0087] The chip transport mechanism detects whether moisture intrusion occurs in the chip batch transport area and the sensitive chip. If so, the control element preheating mechanism is started, and the interference area between the abnormal moisture migration area and the dynamic transport area of ​​the maximum susceptible range of the moisture sensitive chip is obtained;

[0088] Based on the absolute humidity value, the environmental index of the water vapor content in the abnormal moisture migration area is calculated in the moisture migration simulation field to obtain the water vapor content density in the abnormal moisture migration area;

[0089] Based on the architecture information of the moisture-sensitive chip, the circuit structure of the moisture-sensitive chip is drawn, and a chip architecture knowledge graph is obtained, the circuit structure and the water vapor content density are identified in the chip architecture knowledge graph, and a vulnerability index of welding of the moisture-sensitive chip under the water vapor content density condition is obtained;

[0090] Preset an integral interval according to the vulnerability index, and generate a plurality of random sampling points by randomly extracting a probability density function within the interference area to ensure that the integral of the randomly extracted probability density function is within the integral interval;

[0091] Calculate the average value of the probability density function of each random sampling point, and add the average values ​​of all random sampling points to obtain an integral approximation, and determine the probability of soldering damage in the soldering process of the moisture-sensitive chip caused by the water vapor content density in the abnormal moisture migration area according to the integral approximation;

[0092] If the solder loss probability is greater than the preset solder loss probability, the chip transport mechanism is controlled to transfer the moisture-sensitive chip to the component preheating mechanism, and the preheating temperature of the component preheating mechanism is controlled based on the solder loss probability to evaporate and remove moisture on the surface of the chip affected by moisture.

[0093] It should be noted that when it is detected that the chip is invaded by air humidity and moisture, measures should be taken to preheat it to remove water vapor components. However, due to the momentary migration of water vapor components in the air, the water vapor content density in the abnormal moisture migration area changes all the time, that is, the water vapor content density at different time points is not consistent. Therefore, the chips transported at different times are affected by the water vapor humidity in the abnormal moisture migration area to different degrees, resulting in different amounts of water vapor components attached to the chip. If the water vapor component is less or so little that it can be ignored, it will not affect the welding quality of the chip, and there is no need to preheat it. This method can further improve the welding efficiency. Therefore, the method identifies the degree to which the circuit structure of the moisture-sensitive chip is susceptible to damage by the water vapor content density in the abnormal moisture migration area, that is, the vulnerability index of the moisture-sensitive chip when it is welded under the water vapor content density condition; then, based on the vulnerability index, random sampling of the probability density function is performed on the interference area between the abnormal moisture migration area and the dynamic transport area of ​​the maximum susceptible range of the moisture-sensitive chip, so as to calculate the probability of soldering failure that the ambient humidity in the abnormal moisture migration area causes damage to the moisture-invasion sensitive chip during the soldering process or water vapor explosion failure. If the soldering failure probability is less than the preset soldering failure probability, it means that the water vapor content in the abnormal moisture migration area makes the damage effect of the moisture-invasion sensitive chip small or even negligible, so it is not necessary to transport it to the component preheating mechanism; otherwise, it means that the moisture-invasion sensitive chip has a great probability of being damaged or water vapor exploded during the soldering process in the water vapor content in the abnormal moisture migration area, which is likely to cause damage to the overall chip package, so it is necessary to further control the chip transportation mechanism to transport the moisture-invasion sensitive chip to the component preheating mechanism to perform preheating and removal of the humidity and water vapor components. This method can accurately determine and control the preheating of chips attacked by moisture, thereby reducing the damage caused by the influence of environmental humidity and moisture during chip packaging and welding, and effectively reducing the scrap rate of chip packaging and welding.

[0094] More specifically, the step S108 includes the following steps:

[0095] Obtain a designated soldering area on a PCB circuit board reserved for soldering a predetermined chip, and obtain a circuit control architecture and an expected control logic effect of the designated soldering area on the PCB circuit board;

[0096] Kirchhoff's law is introduced, and an accurate circuit control topology map of the specified welding area is constructed in Kirchhoff's law based on the circuit control architecture, and the chip control nodes are extracted through the accurate circuit control topology map;

[0097] Obtain a global circuit control topology of a PCB circuit board, divide the global circuit control topology into a plurality of sub-topology regions, obtain model information of each circuit component, and perform a relaxation planning of a steady-state control voltage integer solution of each circuit component in each sub-topology region using the expected control logic effect and the model information as constraint criteria to obtain an integer programming model;

[0098] Preset an integer solution for a non-steady-state control voltage, obtain a cutting plane of a non-integer constraint contained in an integer programming model, add the cutting plane to a constraint set of the integer programming model and solve it to obtain a new integer solution for a steady-state control voltage;

[0099] If the new steady-state control voltage integer solution still contains a non-integer solution threshold, the cutting plane adding and solving steps are repeated until the non-integer solution threshold is contained, so as to iteratively approximate the steady-state control voltage integer solution of the expected control logic effect, obtain the sub-steady-state control voltage of each circuit component in each sub-topology area, and calculate the expected steady-state voltage value corresponding to each chip control node by combining the sub-steady-state control voltages of all circuit components;

[0100] Creating a control simulation model of the circuit control architecture, using the control simulation model to simulate the circuit control after a given chip is welded in a designated welding area, generating a circuit simulation control result, and obtaining an actual steady-state voltage value of each chip control node through the circuit simulation control result;

[0101] Obtaining a package type of a given chip, calculating a steady-state drift difference between the expected steady-state voltage value and the actual steady-state voltage value, formulating and improving a welding process for the package type according to the steady-state drift difference, and generating a package welding process plan;

[0102] The pin shape, pin number and pin spacing of a given chip are obtained, and the subordinate welding equipment is obtained according to the pin shape and the package welding process plan. The welding schedule of each subordinate welding equipment for each welding point is planned based on the pin spacing to obtain the welding scheduling network.

[0103] It should be noted that there are many packaging and welding methods in chip production, such as BGA packaging welding, QFN packaging welding and SMD patch welding. Different types of chip production packaging methods have different requirements for the packaging welding process. For example, BGA packaging welding has high requirements for the control of solder balls at the pins. However, the existing chip packaging welding technology is difficult to achieve reasonable and accurate decision-making and optimization of the packaging welding process based on the expected circuit control between the chip and the PCB circuit board during the production packaging welding process, resulting in a large error in the welding of chip pins compared to the welding standard under the expected chip circuit operation. The method first determines the chip control nodes through an accurate circuit control topology diagram of a PCB circuit board in which a designated welding area of ​​a given chip welding is located. These chip control nodes are associated with pin circuits. Then, the expected steady-state voltage value of each circuit component in the global circuit control topology diagram of the PCB circuit board is allocated by integer programming. The non-integer cutting plane of the integer programming model is added to exclude the non-integer solution of the current non-steady-state control voltage output from the feasible solution space and iteratively solve it, thereby improving the allocation accuracy and reliability of the sub-steady-state control voltage output of each circuit component of the PCB circuit board to achieve the expected control logic effect, and reducing the packaging welding error caused by the unstable circuit control voltage of the chip on the PCB circuit board; then, the steady-state drift difference between the expected steady-state voltage value and the actual steady-state voltage value is obtained by means of simulation control. The steady-state drift difference reflects the packaging welding requirements of the packaging type when the circuit control between the chip and the PCB circuit board achieves the expected expected control logic effect, so that the welding process of the packaging type can be formulated and improved according to the steady-state drift difference decision, and the scheduling of the welding equipment can be further planned. This method can be used to analyze and decide on the circuit control of chip and PCB packaging welding and optimize the chip packaging welding process, effectively ensure the rationality and reliability of chip packaging welding, and improve the chip packaging welding quality and efficiency.

[0104] More specifically, in the step S108, the pin shape, pin number and pin spacing of a given chip are obtained, the subordinate welding equipment is obtained according to the pin shape and the package welding process plan, and the welding schedule of each subordinate welding equipment for each welding point is planned based on the pin spacing to obtain a welding scheduling network, which specifically includes the following steps:

[0105] Obtain the pin shape, pin number and pin spacing of a given chip, and also obtain the layout design information of the given chip in the specified soldering area on the PCB circuit board;

[0106] Determine the welding equipment required for each welding point on a given chip according to the packaging welding process scheme and the pin shape, define it as subordinate welding equipment, and obtain the historical welding conditions of each subordinate welding equipment for the given chip;

[0107] The historical welding conditions of each subordinate welding equipment are analyzed based on the number of pins to obtain the welding frequency of each subordinate welding equipment. The singular value decomposition algorithm is introduced to perform singular value decomposition on the matrix corresponding to the welding frequency of each subordinate welding equipment to obtain the singular value matrix of the subordinate welding equipment.

[0108] Based on the singular value matrix of the subordinate welding equipment, M singular values ​​are selected to reconstruct the matrix corresponding to the welding frequency of each subordinate welding equipment to determine the singular weight value of each subordinate welding equipment;

[0109] A frequent singular weight threshold is preset. If the singular weight value does not exceed the frequent singular weight threshold, the pin welding points corresponding to one or more subordinate welding devices of the singular weight value are marked as idle welding points; if the singular weight value exceeds the frequent singular weight threshold, the pin welding points corresponding to one or more subordinate welding devices of the singular weight value are marked as frequent welding points;

[0110] According to the singular weight values ​​of idle welding points and frequent welding points, an idle-frequent welding point layout diagram is fitted in the layout design information and the pin spacing. According to the idle-frequent welding point layout diagram, the welding scheduling network of each subordinate welding equipment is planned, and the welding scheduling network is uploaded to the chip packaging welding control terminal.

[0111] It should be noted that in order to save welding time costs and improve the chip packaging welding efficiency, the welding scheduling of the chip welding equipment can be planned; however, the welding scheduling of the existing chip welding equipment usually requires multiple people to collaborate in analyzing the pin information of the chip to be welded and the layout position of the chip on the PCB circuit board for planning, which greatly increases the output of welding time costs and labor costs, is inefficient, time-consuming and labor-intensive, and the scheduling plan finally analyzed and planned cannot follow the welding frequency of the equipment to execute reasonable scheduling or there is a large scheduling error, which leads to errors and welding deviations in the chip welding equipment during operation, making it difficult to ensure the welding quality of the chip package. To this end, the method performs welding frequency analysis on the working conditions of the subordinate welding equipment required for welding each pin when welding the chip in the past, and singularly decomposes a singular weight value for each subordinate welding equipment, and the singular weight value reflects the scheduling priority of each subordinate welding equipment for the given chip welding, so that each subordinate welding equipment can be assigned an idle or frequently welded label when welding different chip pins and their layout on the PCB circuit board according to the singular weight value, thereby further performing reasonable and stable welding scheduling of the chip welding equipment, improving the chip welding efficiency, replacing the cumbersome steps of traditional manual multi-person collaborative planning and scheduling, saving a lot of manpower and welding time cost output, with high economic benefits, and being able to improve the welding quality and stability of the welding equipment for each pin.

[0112] A second aspect of the present invention provides a chip production and packaging welding device, which is applied to any of the chip production and packaging welding control methods described above, such as Figure 3 , 4 As shown, specifically including:

[0113] A chip loading tray 1011, which is used to transport predetermined chips in batches;

[0114] A chip transportation monitoring module 1012, which includes an industrial camera responsible for capturing image data of a plurality of frames of humidity-sensitive chips transported in a chip batch transportation area in a time sequence;

[0115] Humidity detection module 1013, the humidity detection module includes a humidity sensor and an Internet of Things, which is used to obtain a number of groups of humidity data of the virtual transportation space migration sampling points at the preset uniform time stamp of the moisture sensitive chip;

[0116] A component preheating module 1014, which is used to preheat the moisture-sensitive chip to remove water vapor components on its surface;

[0117] A moisture migration simulation field construction module 1015, which is used to construct a moisture migration simulation field around the susceptible point layout in the chip batch transportation area according to the interpolation migration estimation of humidity data;

[0118] The scheduling network planning module 1016 is responsible for planning the welding scheduling network of each subordinate welding equipment according to the idle-frequent welding point layout diagram.

[0119] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0120] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0121] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0122] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0123] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0124] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A welding control method for chip production packaging, characterized in that: The following steps are involved: S102: Obtain the susceptible point layout of the moisture-sensitive chip in the chip batch transportation area and the detection range and distribution pattern diagram of the humidity sensor, constrain the moisture-sensitive chip to be in the maximum susceptible range in the chip batch transportation area based on the susceptible point layout, detection range and distribution pattern diagram, and perform dynamic transportation calculation to obtain a dynamic transportation model in which the maximum susceptible range of the moisture-sensitive chip is in the chip batch transportation area; S104: Acquire several groups of humidity data of the humidity sensor in the virtual transportation space, interpolate each group of humidity data into the Thiessen polygon of the virtual transportation space for migration estimation, so as to describe the moisture migration simulation field around the layout of susceptible points in the chip batch transportation area, and combine the moisture migration simulation field with the dynamic transportation model interference analysis to obtain the chip humidity detection result; S106: Obtaining the interference area between the abnormal moisture migration area and the maximum susceptible range, calculating the soldering probability of the moisture invasion sensitive chip being damaged due to the abnormal moisture migration area based on the interference area and the vulnerability index of the moisture invasion sensitive chip, and controlling the chip transport mechanism to preheat the moisture invasion sensitive chip according to the soldering probability; S108: Construct an accurate circuit control topology diagram in which the designated welding area of ​​a given chip is located on the PCB circuit board, allocate the steady-state control voltage of each circuit component in the accurate circuit control topology diagram based on the expected control logic effect, obtain the steady-state drift difference of each chip control node, and decide the welding process of the package type according to the steady-state drift difference and plan the welding scheduling of the subordinate welding equipment.

2. A chip production packaging welding control method according to claim 1, characterized in that: The step S102 specifically includes the following steps: Obtain the predetermined welding requirements for chip production and packaging and the chip batch transportation area, and extract several predetermined chips that must be welded on the target PCB substrate based on the predetermined welding requirements; Extracting and comparing the single pin features of all the given chips transported in the chip batch transport area, if the similarity between the single pin feature and any reference pin feature in the standard pin comparison warehouse is greater than a preset similarity, the given chip is marked as a moisture sensitive chip; Acquire architecture information of the moisture-sensitive chip, search in a big data network based on the architecture information to identify susceptible points and susceptible point layout of the moisture-sensitive chip, and preset a migration interference range threshold based on the susceptible point layout; Construct a virtual transport space domain for the chip batch transport area, obtain the detection range of the humidity sensor and the distribution pattern diagram of the humidity sensor in the virtual transport space domain, and set the space humidity perception coverage limit according to the distribution pattern diagram; Taking the susceptible range of the humidity sensitive chip in the virtual transportation space as the termination function, constraining the termination function based on the migration interference range threshold until the constrained termination function approaches the perception coverage blind area covered by the space humidity perception coverage boundary, and obtaining the maximum susceptible range; The image data of several frames of humidity-sensitive chips transported in the chip batch transportation area are obtained by shooting with industrial cameras with preset uniform timestamps, and the preset uniform timestamps are decomposed by STL to decompose them into three parts: trend, seasonality and residual. Based on the image data on several frames, a model of the trend part of the transportation of humidity-sensitive chips in the chip batch transportation area is built to capture the dynamic pattern of the residual part, and a trend-residual initial model is obtained. The timestamp time series of the seasonal part is added to the trend-residual initial model, and finally a dynamic transportation model is generated in which the maximum susceptible range of humidity-sensitive chips is in the chip batch transportation area.

3. A chip production packaging welding control method according to claim 1, characterized in that: The step S104 specifically includes the following steps: Migration sampling points are set based on the layout of susceptible points, and several groups of humidity data detected by humidity sensors at migration sampling points in the virtual transportation space at preset uniform timestamps are obtained through the Internet of Things; The virtual transportation space domain is planned into N sub-domains according to the migration sampling points, a Thiessen polygon is constructed based on each sub-domain, an interpolation point of each humidity data is defined, and the interpolation points of several groups of humidity data detected by each migration sampling point are interpolated on the corresponding Thiessen polygon to be interpolated points; After the interpolation is completed, a new Thiessen polygon will be generated for each Thiessen polygon. The overlapping area between the new Thiessen polygon after the interpolation point is inserted and the surrounding original Thiessen polygons is obtained. The ratio of the corresponding overlapping area of ​​the interpolation point and the adjacent interpolation point to the total area of ​​the new Thiessen polygon is calculated to obtain the overlapping area weight value of each interpolation point. The weighted average of the humidity data of each interpolation point is calculated according to the overlapping area weight value, and the migration estimation value of all interpolation points is output. The moisture migration simulation field around the layout of susceptible points in the chip batch transportation area is constructed by describing the migration estimation value of all interpolation points; Based on the moisture migration simulation field, a moisture migration simulation area corresponding to an absolute humidity greater than a preset absolute humidity is stripped off, which is defined as an abnormal moisture migration area, and a dynamic transport area of ​​the maximum susceptible range of the moisture sensitive chip is obtained through the dynamic transport model; If there is no regional interference between the abnormal moisture migration area and the dynamic transport area, the humidity sensitive chip is calibrated as a healthy humidity sensitive chip; if there is regional interference between the abnormal moisture migration area and the dynamic transport area, the humidity sensitive chip is calibrated as a moisture invasion sensitive chip, and the chip humidity detection result is obtained.

4. A chip production packaging welding control method according to claim 1, characterized in that: The step S106 specifically includes the following steps: The chip transport mechanism detects whether moisture intrusion occurs in the chip batch transport area and the sensitive chip. If so, the control element preheating mechanism is started, and the interference area between the abnormal moisture migration area and the dynamic transport area of ​​the maximum susceptible range of the moisture sensitive chip is obtained; Based on the absolute humidity value, the environmental index of the water vapor content in the abnormal moisture migration area is calculated in the moisture migration simulation field to obtain the water vapor content density in the abnormal moisture migration area; Based on the architecture information of the moisture-sensitive chip, the circuit structure of the moisture-sensitive chip is drawn, and a chip architecture knowledge graph is obtained, the circuit structure and the water vapor content density are identified in the chip architecture knowledge graph, and a vulnerability index of welding of the moisture-sensitive chip under the water vapor content density condition is obtained; Preset an integral interval according to the vulnerability index, and generate multiple random sampling points by randomly extracting a probability density function within the interference area to ensure that the integral of the randomly extracted probability density function is within the integral interval; Calculate the average value of the probability density function of each random sampling point, and add the average values ​​of all random sampling points to obtain an integral approximation, and determine the probability of soldering damage in the soldering process of the moisture-sensitive chip caused by the water vapor content density in the abnormal moisture migration area according to the integral approximation; If the solder loss probability is greater than the preset solder loss probability, the chip transport mechanism is controlled to transfer the moisture-sensitive chip to the component preheating mechanism, and the preheating temperature of the component preheating mechanism is controlled based on the solder loss probability to evaporate and remove moisture on the surface of the chip affected by moisture.

5. The welding control method for chip production and packaging according to claim 1, characterized in that: The step S108 specifically includes the following steps: Obtain a designated soldering area on a PCB circuit board reserved for soldering a predetermined chip, and obtain a circuit control architecture and an expected control logic effect of the designated soldering area on the PCB circuit board; Kirchhoff's law is introduced, and an accurate circuit control topology map of the specified welding area is constructed in Kirchhoff's law based on the circuit control architecture, and the chip control nodes are extracted through the accurate circuit control topology map; Obtain a global circuit control topology of a PCB circuit board, divide the global circuit control topology into a plurality of sub-topology regions, obtain model information of each circuit component, and perform a relaxation planning of a steady-state control voltage integer solution of each circuit component in each sub-topology region using the expected control logic effect and the model information as constraint criteria to obtain an integer programming model; Preset an integer solution for a non-steady-state control voltage, obtain a cutting plane of a non-integer constraint contained in an integer programming model, add the cutting plane to a constraint set of the integer programming model and solve it to obtain a new integer solution for a steady-state control voltage; If the new steady-state control voltage integer solution still contains a non-integer solution threshold, the cutting plane adding and solving steps are repeated until the non-integer solution threshold is contained, so as to iteratively approximate the steady-state control voltage integer solution of the expected control logic effect, obtain the sub-steady-state control voltage of each circuit component in each sub-topology area, and calculate the expected steady-state voltage value corresponding to each chip control node by combining the sub-steady-state control voltages of all circuit components; Creating a control simulation model of the circuit control architecture, using the control simulation model to simulate the circuit control after a given chip is welded in a designated welding area, generating a circuit simulation control result, and obtaining an actual steady-state voltage value of each chip control node through the circuit simulation control result; Obtaining a package type of a given chip, calculating a steady-state drift difference between the expected steady-state voltage value and the actual steady-state voltage value, formulating and improving a welding process for the package type according to the steady-state drift difference, and generating a package welding process plan; The pin shape, pin number and pin spacing of a given chip are obtained, and the subordinate welding equipment is obtained according to the pin shape and the package welding process plan. The welding schedule of each subordinate welding equipment for each welding point is planned based on the pin spacing to obtain the welding scheduling network.

6. A chip production packaging welding control method according to claim 5, characterized in that: In the step S108, the pin shape, pin number and pin spacing of a given chip are obtained, the subordinate welding equipment is obtained according to the pin shape and the package welding process plan, and the welding schedule of each subordinate welding equipment for each welding point is planned based on the pin spacing to obtain a welding scheduling network, which specifically includes the following steps: Obtain the pin shape, pin number and pin spacing of a given chip, and also obtain the layout design information of the given chip in the specified soldering area on the PCB circuit board; Determine the welding equipment required for each welding point on a given chip according to the packaging welding process scheme and the pin shape, define it as subordinate welding equipment, and obtain the historical welding conditions of each subordinate welding equipment for the given chip; The historical welding conditions of each subordinate welding equipment are analyzed based on the number of pins to obtain the welding frequency of each subordinate welding equipment. The singular value decomposition algorithm is introduced to perform singular value decomposition on the matrix corresponding to the welding frequency of each subordinate welding equipment to obtain the singular value matrix of the subordinate welding equipment. Based on the singular value matrix of the subordinate welding equipment, M singular values ​​are selected to reconstruct the matrix corresponding to the welding frequency of each subordinate welding equipment to determine the singular weight value of each subordinate welding equipment; A frequent singular weight threshold is preset. If the singular weight value does not exceed the frequent singular weight threshold, the pin welding points corresponding to one or more subordinate welding devices of the singular weight value are marked as idle welding points; if the singular weight value exceeds the frequent singular weight threshold, the pin welding points corresponding to one or more subordinate welding devices of the singular weight value are marked as frequent welding points; According to the singular weight values ​​of idle welding points and frequent welding points, an idle-frequent welding point layout diagram is fitted in the layout design information and the pin spacing. According to the idle-frequent welding point layout diagram, the welding scheduling network of each subordinate welding equipment is planned, and the welding scheduling network is uploaded to the chip packaging welding control terminal.

7. A chip production and packaging welding device, applied to a chip production and packaging welding control method according to any one of claims 1 to 6, characterized in that: Specifically include: A chip loading tray, wherein the chip loading tray is used to transport predetermined chips in batches; A chip transportation monitoring module, wherein the chip transportation monitoring module includes an industrial camera, which is responsible for capturing image data of several frames of humidity-sensitive chips transported in a chip batch transportation area in a time sequence; A humidity detection module, the humidity detection module includes a humidity sensor and an Internet of Things, and is used to obtain a plurality of groups of humidity data of a virtual transport space migration sampling point at a preset uniform time stamp of a moisture sensitive chip; A component preheating module, the component preheating module is used to preheat the moisture-sensitive chip to remove water vapor components on its surface; A moisture migration simulation field construction module, wherein the moisture migration simulation field construction module is used to construct a moisture migration simulation field around the susceptible point layout in the chip batch transportation area according to the interpolation migration estimate of the humidity data; The scheduling network planning module is responsible for planning the welding scheduling network of each subordinate welding equipment according to the idle-frequent welding point layout diagram.

Images