PCBA reliability prediction method, device, storage medium and program product

Abstract

The present application discloses a PCBA reliability prediction method, device, storage medium and program product, which relate to the field of printed circuit board technology. The present application constructs simulation sub-models of temperature field, mechanical field and electric field through the measured temperature data of the PCBA sample, and performs coupling processing, which can more accurately simulate the multi-physical field distribution of the PCBA in the actual working environment. Based on this, simulation analysis can be performed for various failure modes to obtain multi-physical field distribution data under different failure modes. By analyzing the multi-physical field distribution data, the reliability of the PCBA can be predicted. This prediction capability can help technicians identify possible failure risks in advance, so as to take preventive measures and extend the service life of the product. In addition, fault prediction and reliability analysis through simulation models can reduce the number and scope of physical tests, thereby saving development costs and time.

Description

Technical Field

[0001] The present application relates to the technical field of printed circuit boards, and in particular to a PCBA reliability prediction method, device, storage medium, and program product. Background Art

[0002] Temperature detection is particularly important in the production and manufacturing process of printed circuit board assemblies (PCBA).

[0003] In related technologies, a single sensor, thermometer, and static threshold are used to determine whether the temperature of a PCBA is abnormal. This makes it difficult to capture abnormal problems under complex working conditions, resulting in low temperature detection accuracy. Summary of the Invention

[0004] The present application provides a PCBA reliability prediction method, device, storage medium, and program product to at least solve the problem of low temperature detection accuracy in related technologies.

[0005] This application provides a PCBA reliability prediction method, including:

[0006] Collect the temperature of the printed circuit board assembly sample to obtain the measured temperature data;

[0007] According to the measured temperature data, the temperature field simulation sub-model, mechanical field simulation sub-model and electric field simulation sub-model corresponding to the sample are constructed respectively;

[0008] Couple the temperature field simulation sub-model, the mechanical field simulation sub-model, and the electric field simulation sub-model to obtain a multi-physics field simulation model;

[0009] Simulate multiple failure modes using a multi-physics field simulation model to obtain multi-physics field distribution data corresponding to each failure mode;

[0010] According to the multi-physical field distribution data corresponding to each failure mode, reliability prediction is performed on the sample to obtain reliability-related information.

[0011] This application also provides a PCBA reliability prediction device, comprising:

[0012] The acquisition module is used to collect the temperature of the printed circuit board assembly sample to obtain the measured temperature data;

[0013] A construction module is used to construct a temperature field simulation sub-model, a mechanical field simulation sub-model and an electric field simulation sub-model corresponding to the sample according to the measured temperature data;

[0014] A coupling module is used to couple the temperature field simulation sub-model, the mechanical field simulation sub-model and the electric field simulation sub-model to obtain a multi-physics field simulation model;

[0015] A simulation module is used to simulate multiple failure modes using a multi-physics field simulation model to obtain multi-physics field distribution data corresponding to each failure mode;

[0016] The prediction module is used to perform reliability prediction on the sample according to the multi-physical field distribution data corresponding to each failure mode to obtain reliability-related information.

[0017] The present application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for implementing the steps of any of the above-mentioned PCBA reliability prediction methods when executing the computer program.

[0018] The present application also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned PCBA reliability prediction methods are implemented.

[0019] The present application also provides a computer program product, including a computer program, which implements the steps of any of the above-mentioned PCBA reliability prediction methods when executed by a processor.

[0020] This application constructs simulation sub-models of temperature field, mechanical field and electric field through the measured temperature data of the PCBA sample, and performs coupling processing, which can more accurately simulate the multi-physics field distribution of the PCBA in the actual working environment. This precise simulation helps to identify potential failure modes. Based on this, simulation analysis can be performed for multiple failure modes to obtain multi-physics field distribution data under different failure modes. By analyzing the multi-physics field distribution data, the reliability of the PCBA can be predicted. This predictive ability can help technicians identify possible failure risks in advance, so as to take preventive measures and extend the service life of the product. In addition, fault prediction and reliability analysis through simulation models can reduce the number and scope of physical tests, thereby saving development costs and time. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0022] Figure 1 A schematic diagram of a system architecture provided in an embodiment of the present application;

[0023] Figure 2 A schematic diagram of a process flow of a PCBA reliability prediction method provided in an embodiment of the present application;

[0024] Figure 3 A schematic flow chart of another PCBA reliability prediction method provided in an embodiment of the present application;

[0025] Figure 4 A flowchart of an anomaly detection process provided in an embodiment of the present application;

[0026] Figure 5 A schematic structural diagram of a PCBA reliability prediction device provided in an embodiment of the present application;

[0027] Figure 6 This is a schematic diagram of the structure of an electronic device provided in this application. DETAILED DESCRIPTION

[0028] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0029] It should be noted that, in the description of this application, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0030] First, some of the terms involved in this application are explained:

[0031] OPC-UA protocol: The OPC Unified Architecture (UA) protocol. OPC originates from "OLE for Process Control," which means OLE for process control. OLE (Object Linking and Embedding) is a software technology for sharing data and functionality between different applications.

[0032] LSTM-GRU hybrid neural network: A neural network constructed by combining two different types of recurrent neural network (RNN) structures: the Long Short-Term Memory network (LSTM) and the Gated Recurrent Unit (GRU).

[0033] PINN network: Physics-Informed Neural Networks.

[0034] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0035] In conjunction with the specific application environment architecture or specific hardware architecture on which the execution of the PCBA reliability prediction method depends, the specific application environment architecture or specific hardware architecture is described herein.

[0036] Figure 1 This is a schematic diagram of a system architecture provided by an embodiment of the present application. Figure 1 The system architecture includes a PCBA template 10, a temperature sensor array 20, a data acquisition unit 30 and an electronic device 40.

[0037] The PCBA template 10 is an additional template and is not a PCBA in the actual production environment. The PCBA template 10 is used to simulate the actual production process of the PCBA. In order to detect the temperature of the PCBA template 10, a temperature sensor array 20 is provided for the PCBA template 10.

[0038] Optionally, a temperature sensor array 20 is provided in a key thermally sensitive area of ​​the PCBA template 10. The key thermally sensitive area includes at least one of a ball grid array (BGA) area and a device surface. For example, the BGA area is where a central processing unit (CPU) or a graphics processing unit (GPU) is located. The device includes at least one of a power device and a connector. For example, the power device may be a metal oxide semiconductor (MOS) transistor, an inductor, or at least one of other possible devices.

[0039] The temperature sensor array 20 is a multimodal sensor array, that is, it includes a plurality of different types of temperature sensors. Optionally, the temperature sensor array 20 includes at least one of a thermocouple array and a distributed optical fiber temperature measurement unit.

[0040] Optionally, a thermocouple array is arranged in the BGA area. The thermocouple array includes multiple thermocouple monitoring points. The node density and sampling frequency of the thermocouple array can be set according to actual needs, and this embodiment does not limit this. For example, the node density is 4 monitoring points per square centimeter, and the sampling frequency is 200 Hz. For example, the thermocouple array can be installed by using flexible thin film thermocouples with a thickness of ≤0.1 mm, adhered to the component surface with thermally conductive silicone to avoid welding or drilling.

[0041] Optionally, a distributed fiber optic temperature measurement unit is mounted on the device surface. The spatial resolution of the distributed fiber optic temperature measurement unit can be set based on actual needs and is not limited in this embodiment. For example, the spatial resolution can be 0.5 mm. Spatial resolution refers to the minimum distance between two adjacent temperature anomalies (or temperature change points) that can be distinguished within the distributed fiber optic temperature measurement unit. For example, the distributed fiber optic temperature measurement unit can be installed by embedding a fiber optic ribbon into a slot on the edge of the PCBA template 10 or securing it with high-temperature-resistant tape. This utilizes the principle of light scattering to monitor temperature without disrupting the structure of the PCBA template 10.

[0042] Optionally, a temperature sensor array 20 can be set at specific positions around the PCBA template 10 to perform a global temperature scan of the PCBA template 10. Accordingly, the temperature sensor array 20 also includes an infrared thermal imaging unit. Exemplarily, the infrared thermal imaging unit can be set on a fixed bracket perpendicular to the top of the PCBA template 10, and needs to maintain a suitable distance from the PCBA template 10 to ensure that the lens field of view covers the entire template surface. Among them, the wavelength range and frame rate of the infrared thermal imaging unit can be set according to actual needs, and this embodiment does not limit this. For example, the wavelength range can be 8-14 microns (μm), and the frame rate is not less than 30 frames per second (fps) to ensure real-time performance.

[0043] The present application realizes the real-time acquisition of PCBA temperature field through a multi-modal temperature sensor array 20 and a non-destructive installation method, and has a three-dimensional monitoring capability: the infrared thermal imaging unit (surface coverage), the thermocouple array (single point precision), and the distributed optical fiber (path continuity) form a "surface-point-line" collaborative monitoring network. Optionally, the temperature sensor array 20 adopts a star topology and is connected to the data acquisition unit 30 through a controller area network (CAN) bus to ensure the real-time transmission of data. The star topology includes a central node, and all other nodes are connected to this central node. For example, each thermocouple monitoring point is connected to the central node as a node after being integrated through the array, an entire optical fiber of the distributed optical fiber temperature measurement unit is connected to the central node as a whole node, and the infrared thermal imaging unit is connected to the central node as a node.

[0044] Optionally, the data acquisition unit 30 uses Precision Time Protocol (PTP) to achieve time synchronization of the temperature sensor array 20. For example, the clock deviation is controlled within ±1 microsecond (μs) to achieve μs-level time synchronization.

[0045] The data acquisition unit 30 is used to perform denoising on the temperature signal output by the temperature sensor array 20 to obtain a denoised temperature signal and to send the denoised temperature signal to the electronic device 40 .

[0046] The data acquisition unit 30 integrates an adaptive filtering algorithm. Optionally, the adaptive filtering algorithm is a fusion algorithm of wavelet transform and Kalman filtering. Accordingly, the data acquisition unit 30 can be used to remove high-frequency noise (such as electromagnetic interference) in the temperature signal collected by the thermocouple array through wavelet transform. It can also be used to dynamically compensate for low-frequency drift (such as line impedance error) in the temperature signal collected by the distributed fiber optic temperature measurement unit through Kalman filtering. It can also be used to comprehensively process noise (such as pixel offset) in the temperature signal collected by the infrared thermal imaging unit through wavelet transform and Kalman filtering fusion algorithm. Optionally, the wavelet transform uses the db4 wavelet basis.

[0047] Optionally, the data acquisition unit 30 can be any hardware module with data processing and transmission functions, which is not limited in this application. For example, the data acquisition unit 30 can be a CPU, or other general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0048] This application uses a multimodal sensor array to achieve an organic combination of macroscopic global scanning and microscopic local monitoring. Through a collaborative detection mechanism, the temperature detection accuracy is improved, ensuring the accurate capture of small thermal anomalies. Paired with a wavelet transform-Kalman filter fusion algorithm, it effectively suppresses environmental electromagnetic interference and significantly enhances the detection reliability of weak thermal signals.

[0049] The data acquisition unit 30 is also in communication with the electronic device 40 . The data acquisition unit 30 is also used to convert the processed temperature signal into temperature data and send the temperature data to the electronic device 40 .

[0050] The electronic device 40 is used to execute the PCBA reliability prediction method provided herein, modeling the PCBA template 10 to obtain a multi-physics simulation model. The multi-physics simulation model can be considered a digital twin of the PCBA template 10. It can be seen that the data acquisition unit 30 establishes a channel that enables real-time data transmission between the PCBA template 10 and its digital twin.

[0051] Alternatively, the electronic device 40 is intended to be various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers.

[0052] The PCBA reliability prediction method provided in the present application is performed by a PCBA reliability prediction device, which is disposed in an electronic device, such as the electronic device 40 described above.

[0053] Figure 2 A flow chart of a PCBA reliability prediction method provided in an embodiment of the present application is shown as follows: Figure 2 As shown, the PCBA reliability prediction method includes:

[0054] S201: collecting temperature of the PCBA sample to obtain measured temperature data.

[0055] The PCBA sample is an additional sample and is not the PCBA in the actual production environment.

[0056] This embodiment simulates the actual production process of PCBA through a PCBA sample, and detects the temperature of the PCBA sample in real time to obtain measured temperature data.

[0057] The measured temperature data includes the measured temperatures of multiple monitoring points on the PCBA sample.

[0058] S202: Based on the measured temperature data, a temperature field simulation sub-model, a mechanical field simulation sub-model, and an electric field simulation sub-model corresponding to the PCBA sample are respectively constructed.

[0059] The temperature field simulation sub-model is used to simulate the temperature changes of the PCBA sample.

[0060] The mechanical field simulation sub-model is used to simulate the stress changes of the PCBA sample.

[0061] The electric field simulation sub-model is used to simulate the current density changes of the PCBA sample.

[0062] Simulation software can be used to construct a temperature field simulation sub-model, a mechanical field simulation sub-model, and an electric field simulation sub-model corresponding to the PCBA template. The simulation software can be selected according to actual needs and is not limited in this embodiment.

[0063] S203: performing coupling processing on the temperature field simulation sub-model, the mechanical field simulation sub-model, and the electric field simulation sub-model to obtain a multi-physics field simulation model.

[0064] The multi-physics simulation model can be considered a digital twin of the PCBA prototype. During actual PCBA production, the temperature field changes, and the mechanical and electric fields also change as a result. Therefore, the temperature field simulation sub-model, mechanical field simulation sub-model, and electric field simulation sub-model of the PCBA prototype can be coupled to achieve multi-physics coupling, enabling comprehensive analysis of the PCBA prototype based on the multi-physics simulation model.

[0065] S204: Simulating multiple failure modes respectively through a multi-physics field simulation model to obtain multi-physics field distribution data corresponding to each failure mode.

[0066] Failure modes are used to describe possible failures of PCBAs in actual production scenarios. The specific content and number of failure modes can be set based on actual needs and are not limited in this application.

[0067] Multi-physics field distribution data includes temperature field distribution data, stress distribution data, and current density distribution data.

[0068] In this embodiment, a multi-physics field simulation model is constructed in S203, and then a fault simulation application is performed based on the constructed model in S204.

[0069] S205: Perform reliability prediction on the PCBA sample according to the multi-physics field distribution data corresponding to each failure mode to obtain reliability-related information.

[0070] Among them, the reliability-related information is used to describe the reliability of the PCBA sample.

[0071] This application uses the measured temperature data of the PCBA sample to construct simulation sub-models of the temperature field, mechanical field, and electric field, and performs coupling processing, which can more accurately simulate the multi-physics field distribution of the PCBA in the actual working environment. This precise simulation helps to identify potential failure modes. Based on this, simulation analysis can be performed for multiple failure modes to obtain multi-physics field distribution data under different failure modes. By analyzing the multi-physics field distribution data, the reliability of the PCBA can be predicted. This predictive ability can help technicians identify possible failure risks in advance, so that preventive measures can be taken to extend the service life of the product. In addition, fault simulation and reliability prediction through simulation models can reduce the number and scope of physical tests, thereby saving development costs and time.

[0072] Figure 3 A flow chart of another PCBA reliability prediction method provided in an embodiment of the present application is shown as follows: Figure 3 As shown, the PCBA reliability prediction method includes:

[0073] S301: Temperature of the PCBA sample is collected through a multi-modal sensor array to obtain raw temperature data.

[0074] The multimodal sensor array includes a thermocouple array, a distributed fiber optic temperature measurement unit, and an infrared thermal imaging unit. The specific implementation process of temperature acquisition on a PCBA sample using the multimodal sensor array is described in the description of the temperature sensor array 20 in the system architecture above and will not be repeated here.

[0075] S302: De-noising the original temperature data to obtain measured temperature data.

[0076] In some embodiments, the electronic device can be directly connected to the multimodal sensor array to obtain raw temperature data and perform denoising on the raw temperature data to obtain measured temperature data. The specific implementation process of the denoising process is similar to the denoising process of the data acquisition unit 30 in the above-mentioned system architecture and will not be repeated here.

[0077] S301-S302 is a method for collecting temperature of PCBA samples to obtain measured temperature data. The raw temperature data collected by the multimodal sensor array is more comprehensive. By denoising the raw temperature data, more accurate measured temperature data can be obtained.

[0078] In other embodiments, the electronic device may also be communicatively connected to a data acquisition unit, which acquires raw temperature data collected by the multimodal sensor array, performs denoising on the raw temperature data, obtains measured temperature data, and transmits the measured temperature data to the electronic device. Accordingly, steps S301-S302 may be replaced by: acquiring measured temperature data sent by the data acquisition unit, obtained by performing temperature acquisition and denoising on the PCBA sample. The specific implementation process is described in the relevant description of data acquisition unit 30 in the system architecture above and will not be repeated here.

[0079] S303: Construct a temperature field simulation sub-model corresponding to the PCBA sample based on the measured temperature data.

[0080] In an optional implementation, the specific implementation process of S303 includes S3031-S3032:

[0081] S3031: Construct heat conduction sub-model, convection heat dissipation sub-model and radiation heat transfer sub-model for PCBA sample.

[0082] S3032: Couple the heat conduction sub-model, the convection heat dissipation sub-model, and the radiation heat transfer sub-model to obtain a temperature field simulation sub-model.

[0083] Optionally, S3031 is specifically implemented as follows:

[0084] Use simulation software to create a base structural model of the PCBA, including multiple components. Initial material parameters are set for each component. The base structural model is a 3D model, and the component material parameters represent the material properties. For example, multiple components may include the PCB substrate, BGA, capacitors, resistors, connectors, and other components. Material parameters include thermal conductivity (also known as thermal conductivity), density, specific heat capacity, convective heat transfer coefficient, emissivity, and other possible parameters.

[0085] A thermal conductivity sub-model is constructed for the original structural model. This sub-model is used to simulate heat conduction between the PCBA template and components. Specifically, the sub-model is expressed as a heat conduction equation, which can be established based on Fourier's law. The material parameters required for modeling include the material's thermal conductivity, density, and specific heat capacity. For example, the thermal conductivity coefficients for copper foil are set to 401 W / (m·K) and the thermal conductivity for FR-4 substrates are set to 0.3 W / (m·K).

[0086] For example, the heat conduction equation is , where ρ is the material density, c is the specific heat capacity, T is the temperature field distribution, t is the time, k is the thermal conductivity, and Q is the internal heat source.

[0087] A convection cooling sub-model was constructed for the original structural model. This sub-model simulates heat exchange between air and the PCBA sample surface, a process known as forced convection. Specifically, the k-ε turbulence model was used. The material parameters required for this modeling included the convection heat transfer coefficient. Furthermore, the fan speed parameters required for this modeling were dynamically updated using measured data (e.g., the real-time wind speed during device operation).

[0088] A radiation heat transfer submodel was constructed for the original structural model. This submodel simulates the radiative heat transfer between the component and the housing. Specifically, this model is based on the Stefan-Boltzmann law. Parameters required for this model include emissivity, which is determined based on the component's surface treatment (e.g., plating or oxide layer) and can be obtained from material handbooks or experimental measurements.

[0089] When a PCBA prototype is operating, internal components transfer heat to the outer casing through conduction (the heat conduction sub-model). The outer casing dissipates heat through air convection (the convection heat dissipation sub-model), while simultaneously radiating heat to the surrounding environment (the radiation heat transfer sub-model). Modeling these three heat transfer processes separately can lead to errors, but coupling allows for a coordinated description of these three heat transfer processes.

[0090] The resulting heat balance equation, formed by coupling the three heat transfer processes, is strongly nonlinear. This means the temperature field simulation submodel can be represented as a nonlinear heat balance equation. The nonlinear heat balance equation, f(T) = 0, describes the physical relationship between the temperature field distribution T and the material parameters.

[0091] Optionally, in order to improve the simulation accuracy of the temperature field simulation sub-model, the temperature field simulation sub-model may be reversely optimized. Accordingly, after S3032, the specific implementation process of S303 further includes S3033-S3037:

[0092] S3033: Use the Newton-Raphson algorithm to solve the nonlinear thermal equilibrium equation to obtain the temperature field distribution data under the current material parameters.

[0093] Among them, the initial material parameters of the nonlinear heat balance equation are known, and the equation is a nonlinear equation about the temperature field distribution T, which can be directly solved by the Newton-Raphson algorithm.

[0094] The temperature field distribution data includes the simulated temperature of each node in the simulation grid. The simulation grid is obtained by discretizing the PCBA three-dimensional model into a finite number of tiny units (such as hexahedrons and tetrahedrons).

[0095] S3034: Determine a residual matrix based on the measured temperature data and the temperature field distribution data of the PCBA sample.

[0096] The measured temperature data is collected by a multimodal sensor array. To facilitate calculation, the temperature data collected by a thermocouple array and a distributed optical fiber temperature measurement unit can be used.

[0097] The measured temperature data includes the measured temperatures at multiple monitoring points, and the temperature field distribution data includes the simulated temperatures at each node in the simulation grid. Therefore, the multiple monitoring points and the multiple nodes in the simulation grid can be spatially aligned before calculating the residual. The alignment method can refer to the alignment method in the related art and is not described in detail in this embodiment.

[0098] The residual matrix includes the residuals of each point after alignment.

[0099] S3035: Determine the residual root mean square error (RMSE) according to the residual matrix.

[0100] S3036: If the residual root mean square error is greater than the preset error value, reverse optimization is performed on the material parameters in the temperature field simulation sub-model to obtain an optimized temperature field simulation sub-model.

[0101] The preset error value can be set according to actual needs, and this embodiment does not limit this. For example, if the temperature unit is ° C., the preset error value can be 2, 2.5 or 3.

[0102] If the residual root mean square error is less than or equal to the preset error value, it means that the error between the measured temperature and the simulated temperature is small, and there is no need to perform reverse optimization on the temperature field simulation sub-model.

[0103] If the residual root mean square error is greater than the preset error value, it means that the error between the measured temperature and the simulated temperature is large, and the accuracy of the temperature field simulation sub-model is not high. Therefore, the temperature field simulation sub-model can be reverse optimized to make the model output closer to the measured temperature by updating the material parameters.

[0104] Optionally, the material parameters may be updated using an optimization algorithm, such as gradient descent or Gauss-Newton method, which is not limited in this embodiment.

[0105] S3037: Repeat the above operation until the residual root mean square error is less than or equal to the preset error value.

[0106] The core of reverse optimization is to minimize the error between the measured and simulated temperatures. By repeating steps S3033-S3036 until the residual root mean square error converges, the material parameters at this point are the optimal solution.

[0107] For example, the iteration tolerance of the Newton-Raphson algorithm can be set to 1×10 -6 or other suitable values.

[0108] In this embodiment, each update of the material parameters requires resolving the temperature field distribution data. The Newton-Raphson algorithm can quickly obtain a high-precision temperature solution due to its quadratic convergence characteristics, providing a reliable gradient calculation basis for parameter optimization.

[0109] It should be noted that the temperature field simulation sub-model can be reverse optimized periodically to ensure the accuracy of the model. For example, the period can be 20 minutes, 30 minutes, 40 minutes, etc., which is not limited in this embodiment.

[0110] By comparing measured temperature data with simulated temperature data, material parameters are updated, forming a dynamic correction mechanism of "simulation-measurement-optimization" to ensure that the model always closely matches actual operating conditions. This allows the temperature field simulation sub-model to automatically update with process changes in real-world PCBA production environments, minimizing long-term fluctuations in simulation accuracy.

[0111] In an optional implementation, the simulation grid corresponding to the PCBA template can also be dynamically subdivided. Accordingly, the method provided by this application also includes:

[0112] The temperature gradient between adjacent nodes in the simulation grid corresponding to the PCBA template is determined based on the temperature field distribution data; if the temperature gradient is greater than the preset gradient value, the grid area where the adjacent nodes are located is meshed to obtain a new simulation grid.

[0113] The temperature field distribution data is obtained by solving S3033. The temperature gradient G is the product of the temperature difference between adjacent nodes and the distance between nodes. The calculation formula is: G = ΔxΔT, where ΔT is the temperature difference between adjacent nodes and Δx is the distance between nodes.

[0114] If the temperature gradient is greater than the preset gradient value, it indicates that the temperature difference between adjacent nodes is large, so the grid can be refined to subdivide the grid. The preset gradient value can be 5°C / mm or other suitable values, which is not limited in this embodiment.

[0115] For example, the initial grid size is set to 1mm×1mm×0.2mm. When the temperature gradient is greater than 5℃ / mm, the grid is automatically encrypted. For example, the original grid side length is refined from 1mm to 0.2mm, and the new grid size is 0.2mm×0.2mm×0.05mm.

[0116] In this embodiment, high-temperature gradient areas (such as solder joints and power device edges) require a denser grid to capture subtle temperature changes, reduce calculation errors caused by sparse grids, and improve local calculation accuracy through dynamic grid subdivision.

[0117] S304: Construct a mechanical field simulation sub-model corresponding to the PCBA sample.

[0118] A mechanical field simulation sub-model is constructed for the original structural model. Temperature changes cause changes in thermal strain, which in turn causes changes in stress. Accordingly, the mechanical field simulation sub-model is expressed in the form of mechanical equations. The mechanical equation is ,in is the stress divergence, is the body force, i is the free index, and j is the dummy index. The stress distribution of the PCBA sample caused by temperature change is simulated through the mechanical equation.

[0119] S305: Construct an electric field simulation sub-model corresponding to the PCBA sample.

[0120] An electric field simulation sub-model is constructed for the original structural model. Temperature changes cause changes in conductivity, which in turn causes changes in current density. Accordingly, the electric field simulation sub-model is expressed in the form of an electric field equation, which is: ,in, is the current density, and , is the conductivity as a function of temperature, is the electric field, , The electric field equation is used to simulate the change in current density of the PCBA sample due to temperature changes.

[0121] S306: performing coupling processing on the temperature field simulation sub-model, the mechanical field simulation sub-model, and the electric field simulation sub-model to obtain a multi-physics field simulation model.

[0122] Specific coupling method: First, perform thermal simulation through the temperature field simulation sub-model to obtain temperature field distribution data; map the temperature field distribution data to the mechanical field simulation sub-model and the electric field simulation sub-model, that is, use the temperature field distribution data as the "temperature field input" of thermal-mechanical coupling and thermal-electric coupling respectively; simulate the mechanical field simulation sub-model and the electric field simulation sub-model respectively to realize the multi-physics field solution of the basic working conditions.

[0123] Most simulation software supports the "field coupling" function, which can automatically complete temperature mapping.

[0124] S307: Simulating multiple failure modes respectively through a multi-physics field simulation model to obtain multi-physics field distribution data corresponding to each failure mode.

[0125] In an optional implementation, the specific implementation process of S307 includes: performing the following operations for each failure mode, respectively, the following operations including:

[0126] Inject the failure mode into the temperature field simulation sub-model in a parameterized form;

[0127] Numerical solution is performed on the temperature field simulation sub-model to obtain temperature field distribution data;

[0128] Using the temperature field distribution data as input, the mechanical field simulation sub-model is driven to perform numerical solutions to obtain stress distribution data;

[0129] Using the temperature field distribution data as input, the electric field simulation sub-model is driven to perform numerical solution to obtain the current density distribution data.

[0130] The current density distribution data includes the current density of each node in the simulation grid, and the stress distribution data includes the stress of each node in the simulation grid.

[0131] Optionally, multiple failure modes are pre-set, and each failure mode is converted into a temperature field boundary condition or material parameter change to obtain a failure parameter. When simulating the failure mode, the corresponding failure parameter is injected into the temperature field simulation sub-model and the temperature field distribution data is solved.

[0132] Adjustable fault parameters are defined for each fault mode. For example, a heat sink detachment corresponds to a 50% reduction in the convection coefficient, which can be achieved by modifying the boundary conditions; another example is component parameter drift, which can be achieved by modifying the temperature coefficient of the resistor / capacitor.

[0133] Optionally, a fault mode can be randomly selected from a variety of fault modes using a preset probability distribution (such as uniform distribution or Poisson distribution), or injected in the order of test scenarios (such as simulating thermal failure first and then mechanical failure) to verify the robustness of the PCBA sample under different fault combinations.

[0134] Exemplarily, the multiple failure modes include at least the following:

[0135] 1. Heat sink falls off: resulting in a decrease in convection coefficient.

[0136] 2. Component overheating and burning: Simulate some key components (such as power devices, CPU, etc.) due to excessive power consumption, poor heat dissipation, etc., which lead to excessive temperature and eventually burn out, thereby affecting the normal operation of the entire PCBA.

[0137] 3. Component parameter drift: The parameters of components such as resistors and capacitors drift due to temperature changes, resulting in changes in circuit performance, signal anomalies, functional failures, and other problems.

[0138] 4. Cold solder joints: During the soldering process, improper soldering technology or inaccurate temperature control can cause cold solder joints. When the temperature changes, the resistance of the cold solder joints will change, which may cause the circuit to be intermittent and affect the stability of the PCBA.

[0139] 5. Solder joint cracking: Under long-term temperature cycling or mechanical stress, solder joints may crack. This can cause the electrical connection between the component and the PCB substrate to deteriorate or even completely disconnect, causing failure.

[0140] 6. Pin short circuit: Short circuit between component pins due to improper welding, foreign matter contamination, etc., causes abnormal current flow and local overheating, which may also affect the normal operation of related circuits.

[0141] This embodiment achieves the re-solution of multiple physical fields under fault conditions by simulating fault modes.

[0142] S308: Perform reliability prediction on the PCBA sample according to the multi-physics field distribution data corresponding to each failure mode to obtain reliability-related information.

[0143] The reliability-related information is used to describe the reliability of the PCBA sample. The reliability-related information includes at least one of signal attenuation information, regional fracture information, and electromigration information. The signal attenuation information indicates the possibility of signal attenuation in a first region of the PCBA sample. The regional fracture information indicates the possibility of fracture in a second region of the PCBA sample. The electromigration information indicates the possibility of electromigration in a third region of the PCBA sample.

[0144] In an optional implementation, the specific implementation process of S308 includes: performing the following operations for each failure mode, the following operations including S3081-S3083:

[0145] S3081: Determine the conductor resistance change data based on the temperature field distribution data after the fault mode injection and the temperature field distribution data before the fault mode injection.

[0146] The wire resistance change data includes the resistance change of each node in the simulation grid.

[0147] Temperature changes directly lead to changes in resistance. For example, rising temperature increases resistance. Accordingly, for each node in the simulation grid, the resistance change at that node is determined based on the temperature change at that node. The temperature change at a node is the difference between the simulated temperature before and after the fault mode is injected.

[0148] The resistance change of the node is given by the formula Calculated. Where ΔR represents the resistance change, is the initial resistance, that is, the resistance value when the temperature is the reference temperature (usually 20℃ or 0℃), ΔT is the temperature change, α is the temperature coefficient, and α=0.0039 / ℃.

[0149] S3082: Determine signal attenuation information based on the wire resistance change data.

[0150] The attenuation of a signal during transmission in a wire is positively correlated with resistance, so an increase in resistance causes signal attenuation. Optionally, S3082 is specifically implemented as follows: if the resistance change of any node in the simulation grid indicates an increase in resistance at that node, then, with the region where the node is located as the first region, generating signal attenuation information indicating that signal attenuation may exist in the first region.

[0151] S3083: Determine regional fracture information based on stress distribution data after fault mode injection and stress distribution data before fault mode injection.

[0152] Rising temperatures create hotspots, affecting the mechanical properties of the material, such as increasing stress and exacerbating the risk of fracture. Optionally, S3083 is implemented as follows: if the stress change at any node in the simulation grid indicates an increase in stress at that node, then the region where the node is located is defined as the second region, and regional fracture information is generated to indicate the potential for fracture in the second region. The stress change at the node is the difference between the stress data before and after the fault mode injection.

[0153] S3084: Determine electromigration information based on the current density distribution data after the fault mode injection and the current density distribution data before the fault mode injection.

[0154] Temperature changes cause changes in conductivity. To satisfy the electric field equation, the current density will adaptively adjust. For example, if the conductivity of a certain area decreases due to temperature increase, the current density in that area will decrease. To maintain current continuity, the current density in adjacent areas must increase accordingly. Areas with concentrated current density (such as solder joint corners and areas where the conductor cross-section changes) are prone to electromigration and are high-risk solder joints that may cause copper foil melting or substrate carbonization.

[0155] Optionally, S3084 is specifically implemented as follows: if the change in current density of any node in the simulation grid indicates an increase in current density at the node, taking the region where the node is located as a third region, and generating electromigration information indicating that electromigration may exist in the third region. The change in current density of the node is the difference between current density data before and after the fault mode is injected.

[0156] Through thermal-electro-mechanical coupling analysis, the impact of faults on signal integrity (such as signal attenuation caused by resistance changes), mechanical reliability (such as solder joint cracking), and electromigration is predicted, achieving reliability prediction and providing an improvement basis for subsequent PCBA production and manufacturing.

[0157] exist Figure 2 or Figure 3 Based on the embodiment shown, anomaly detection can also be performed on the temperature field simulation sub-model. Figure 4 A flowchart of an anomaly detection process provided in an embodiment of the present application is shown as follows: Figure 4 As shown in Figure 2, the anomaly detection process includes:

[0158] S401: Acquire temperature field distribution data output by the temperature field simulation sub-model.

[0159] Optionally, the implementation method of this step is the same as that of S3033 and will not be repeated here.

[0160] S402: Determine whether there is any abnormality in the temperature field distribution data using a preset detection algorithm.

[0161] The preset detection algorithm may be any one of a thermal behavior benchmark model, an isolation forest-autoencoder hybrid algorithm, or a temperature evolution prediction model.

[0162] S403: If an anomaly exists, an anomaly detection result is generated.

[0163] Anomaly detection results indicate anomalies in the temperature field distribution data. Specifically, the results include node location information and temperature anomaly values. A node is a node in the simulation grid with an abnormal temperature. The location information represents the node's coordinates within the simulation grid. The temperature anomaly value is the difference between the node's simulated temperature and a reference temperature, which is determined by a pre-set detection algorithm.

[0164] S404: Match the anomaly detection result with the fault type recorded in the database to obtain the target fault type.

[0165] The database stores a large number of PCBA-related fault types, and each fault type may cause temperature abnormality of the PCBA.

[0166] Optionally, the database is connected to a rule engine, and multiple temperature anomaly association rules are set in the rule engine. The rule engine can then match the anomaly detection results with the fault types recorded in the database according to the temperature anomaly association rules to achieve fault tracing.

[0167] Specifically, the abnormal characteristic parameters are parsed from the abnormality detection results. The abnormal characteristic parameters include location information and temperature anomaly values. Candidate rules are screened from multiple temperature anomaly association rules based on the location information, and then it is determined whether the temperature anomaly value meets the triggering conditions of the candidate rules. If so, the fault type corresponding to the candidate rule is used as the target fault type.

[0168] For example, a temperature anomaly association rule is triggered by a node located in the solder joint area and a temperature anomaly value greater than 8°C, matching the fault type as a poor solder joint. Another temperature anomaly association rule is triggered by a node located in the device core area and a temperature anomaly value greater than 15°C, matching the fault type as poor device heat dissipation. Yet another temperature anomaly association rule is triggered by multiple temperature anomalies (≥3 nodes) and an average temperature anomaly value greater than 6°C, matching the fault type as power supply component aging.

[0169] Optionally, the database also stores data related to product design for manufacturability, including product design specifications, manufacturing process parameters, material properties, cost data, and previous design cases. For example, the database can be at least one of a R&D design library, a manufacturing process problem library, or a Design for Manufacturing (DFM) evaluation database.

[0170] S405: Feedback the abnormality detection result and the target fault type to the manufacturing execution system.

[0171] Optionally, the anomaly detection results and target fault types are fed back to the manufacturing execution system via a data communication protocol in the industrial automation field (such as the OPC-UA protocol).

[0172] By using the abnormal detection results and the matched target fault types as feedback information for the manufacturing execution system, technicians can optimize the PCBA manufacturing process, thus achieving closed-loop feedback optimization of the PCBA production and manufacturing process.

[0173] In some embodiments, the temperature field simulation sub-model can be periodically detected for anomalies, thereby providing timely feedback to the manufacturing execution system. The anomaly detection period can be set according to actual needs and is not limited in this embodiment.

[0174] S403-S405 provide a processing method for abnormal situations. If the temperature field distribution data is determined to be normal by the preset detection algorithm, no processing is performed and the next abnormality detection is waited for.

[0175] In some embodiments, the PCBA design version can be bound to the anomaly detection results based on the blockchain's traceability mechanism to facilitate subsequent tracing of quality issues.

[0176] exist Figure 4 Based on the embodiment shown, S402 includes at least one of the following implementations:

[0177] The first implementation method uses a trained thermal behavior benchmark model to determine whether there are any anomalies in the temperature field distribution data. The thermal behavior benchmark model is used to determine the predicted temperature range based on the PCBA's current operating condition data, thereby predicting the PCBA's current possible temperature range.

[0178] Optionally, a specific implementation process of using the trained thermal behavior benchmark model to determine whether the temperature distribution data is abnormal includes:

[0179] Obtain the current operating condition data of the PCBA sample; input the operating condition data as input data into the thermal behavior benchmark model, and output the predicted temperature range through the thermal behavior benchmark model; determine whether the temperature field distribution data is abnormal based on the predicted temperature range.

[0180] The operating condition data includes at least one of electrical parameters, physical environment parameters, and operating status parameters. Electrical parameters include at least one of component voltage, current, power consumption, and impedance. Physical environment parameters include at least one of temperature field distribution (temperature of each node), humidity, and vibration frequency / amplitude. Operating status parameters include at least one of signal transmission delay, clock frequency stability, and data throughput.

[0181] The predicted temperature range includes the temperature range of each node in the simulation grid. This temperature range encompasses the range within which the node's temperature may fluctuate under the current operating conditions. Accordingly, for each node, if the simulated temperature of that node in the temperature field distribution data falls within the node's temperature range, the node's temperature is normal. Conversely, if the simulated temperature of that node in the temperature field distribution data falls outside the node's temperature range, the node's temperature is abnormal. When determining the abnormal temperature value of a node, the median temperature within the node's temperature range can be used as a reference temperature.

[0182] Optionally, the training process of the thermal behavior benchmark model includes:

[0183] Obtain sample data and label data; the sample data includes historical operating condition data of the PCBA, and the label data includes the measured temperature range under the operating condition represented by the sample operating condition data;

[0184] The labeled data is used as the supervised data, and the initial thermal behavior benchmark model is iteratively trained according to the sample data to obtain a trained thermal behavior benchmark model.

[0185] To improve training accuracy, a large amount of sample data and label data corresponding to the sample data can be obtained. For example, the sample data can be the normal operating data of a PCBA in an actual production environment within 1000 hours (such as collecting a set of operating data every 1 minute or 1 hour), and the label data can be the actual measured temperature range collected by the sensor under the corresponding operating conditions.

[0186] By training the thermal behavior benchmark model, it is possible to predict the accurate temperature range based on the current operating condition data, and then use this temperature range as the threshold range to determine whether the temperature field distribution data is abnormal, with high accuracy.

[0187] Optionally, the model architecture of the thermal behavior benchmark model can be an LSTM-GRU hybrid neural network. Table 1 shows a network architecture of an LSTM-GRU hybrid neural network.

[0188] Table 1

[0189]

[0190] It should be noted that Table 1 is only an example. In actual applications, the model architecture of the thermal behavior benchmark model can also be in other forms, and this application does not limit this.

[0191] The second implementation method is to use the isolation forest-autoencoder hybrid algorithm to determine whether there are abnormalities in the temperature field distribution data.

[0192] Isolation forests use tree structures to quickly identify outliers and are suitable for sparse anomalies in high-dimensional data. Autoencoders capture data distribution characteristics through reconstruction errors and are more sensitive to local and clustered anomalies. The isolation forest-autoencoder hybrid algorithm combines the fast screening capabilities of isolation forests with the reconstruction accuracy of autoencoders.

[0193] Specifically, an isolation forest algorithm is used to perform preliminary anomaly detection on temperature field distribution data, quickly identifying clearly abnormal samples. Samples deemed normal by the isolation forest algorithm are extracted for training an autoencoder, and a multi-layer perceptron autoencoder is constructed to learn the characteristic representation of normal temperature distributions. The reconstruction error of each sample is calculated as the autoencoder's anomaly score. The scores from the isolation forest and autoencoder are weighted and fused to obtain the final anomaly score. Only samples deemed abnormal by both models are marked as abnormal. The sample refers to the simulated temperature of the node. Since the autoencoder outputs a reconstructed temperature value for each node, which reflects the autoencoder's fitting result for "normal temperature," the reconstructed temperature value of the node can be used as a reference temperature when determining abnormal temperature values ​​for a node.

[0194] The autoencoder is trained to minimize reconstruction error. In the temperature field distribution data, nodes that meet both the isolation forest threshold (e.g., score > 0.7) and the autoencoder threshold (e.g., reconstruction error > 3σ) are marked as abnormal. A reconstruction error > 3σ is based on the 3σ principle of the normal distribution.

[0195] The contamination rate of the isolation forest algorithm can be set to 0.1% or other possible values, which are not limited in this application. Setting the contamination rate to 0.1% means that during the detection process, it is expected that only 0.1% of the data points in the data will be determined as outliers.

[0196] This implementation adopts the isolation forest-autoencoder hybrid algorithm, which takes into account both local outlier recognition and global pattern learning.

[0197] The third implementation method uses a trained temperature evolution prediction model to determine whether there are any anomalies in the temperature field distribution data. The temperature evolution prediction model is used to continuously predict the temperature field change information of the PCBA over a period of time.

[0198] Optionally, a specific implementation process of using the trained temperature evolution prediction model to determine whether the temperature field distribution data is abnormal includes:

[0199] The temperature field distribution data and the corresponding load parameters are used as input data and input into the trained temperature evolution prediction model. The temperature evolution prediction model is used to roll out predictions of the temperature field change information at each time point within a preset time period in the future. The presence of anomalies is determined based on the temperature field change information.

[0200] This implementation uses the temperature field distribution data currently simulated by the temperature field simulation sub-model and the current load parameters of the PCBA sample as input data for the temperature evolution prediction model, and outputs temperature field change information at each time point within a preset time period in the future.

[0201] Load parameters refer to various external input conditions or operating state parameters that affect the temperature distribution of the PCBA template. These parameters directly or indirectly cause temperature changes at various nodes on the PCBA and serve as important input variables for the temperature evolution prediction model's rolling prediction. Exemplarily, load parameters include at least one of electrical load parameters, environmental load parameters, and structural load parameters. Electrical load parameters include at least one of the operating current, voltage, power consumption, and power input voltage of each component. Environmental load parameters include at least one of ambient temperature, humidity, heat dissipation conditions (such as air flow rate, forced air cooling / water cooling flow rate), and ambient air pressure. Structural load parameters include at least one of the PCBA template's material properties (such as thermal conductivity and copper foil thickness), component layout density, and packaging method.

[0202] The preset duration can be set according to actual needs, such as 10 minutes, 20 minutes, etc., and this application does not limit this. The time step can also be set according to actual needs, such as 1 second, 2 seconds, etc., and this application does not limit this.

[0203] For the temperature field change information at each time point within a preset time period in the future, the temperature field change information includes the predicted temperature value of each node in the simulation grid at that time point. Accordingly, the operation of determining whether there is an abnormality based on the temperature field change information is specifically implemented as follows: for each node, if the difference between the predicted temperature values ​​of the node at two adjacent time points is greater than the preset difference, it is determined that the node has an abnormality, and the difference is directly determined as the temperature abnormality value of the node; conversely, if the difference between the predicted temperature values ​​of the node at two adjacent time points is less than or equal to the preset difference, it is determined that there is no abnormality at the node. Among them, the preset difference can be set according to actual needs, and this embodiment does not limit this, for example, 2°C, 3°C, etc.

[0204] The temperature evolution prediction model is used to predict the temperature field change information in the future, thus providing a basis for preventive maintenance.

[0205] Optionally, the temperature evolution prediction model training process includes iteratively training the initial temperature evolution prediction model by minimizing a loss function to obtain a trained temperature evolution prediction model. The loss function is determined based on a heat conduction equation and a data fitting error; the data fitting error is the mean square error between the predicted temperature and the sample temperature.

[0206] The sample data required for training is the historical temperature data and load parameters of the PCBA at various time points over a period of time. The historical temperature data includes the measured temperature of each node. The PCBA can be a PCBA in an actual production environment.

[0207] The loss function consists of two parts: a physical constraint term and a data-driven loss. The physical constraint term is the residual of the heat conduction equation, and the data-driven loss is the data fitting error, which is the mean squared error between the predicted temperature and the sample temperature. The sample temperature is the measured temperature.

[0208] For example, the loss function is shown in the following formula:

[0209] L=L data +λL physics

[0210] Among them, L is the model loss, L data is the data-driven loss, L physics is a physical constraint term, λ is a weight coefficient, and λ can be set based on experience, which is not limited in this application.

[0211] By determining the model's loss function based on the heat conduction equation and data fitting error, the trained model can make predictions more accurate.

[0212] This embodiment provides multiple implementation methods for determining whether temperature field distribution data is abnormal. In practical applications, at least one implementation method can be flexibly selected, which is highly flexible.

[0213] Building on the above embodiments, in some embodiments, mixed reality devices (such as the HoloLens 2) can overlay and display different images and information. These include real-time infrared images, whose transparency can be set to 50%; simulated temperature field isotherms, whose intervals can be set to 5°C; and difference cloud maps, where red indicates that the measured value is greater than the simulated value, and blue indicates that the measured value is less than the simulated value. The real-time infrared image is generated based on the temperature data output by the infrared thermal imaging unit. The simulated temperature field isotherms and difference cloud maps are generated based on the temperature field distribution data output by the temperature field simulation sub-model. By leveraging visualization technology, the overlay display of images and information facilitates technical personnel's review, reducing fault location time and thereby improving production and maintenance efficiency.

[0214] Building on the above-mentioned embodiments, in some embodiments, an edge-cloud collaborative computing architecture can also be deployed. For example, data preprocessing and real-time anomaly detection can be performed at the edge, while model training can be performed in the cloud. This edge-cloud collaborative computing architecture achieves low latency for real-time anomaly detection at the edge, fast model training in the cloud, and supports scalability to thousands of nodes.

[0215] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method.

[0216] Figure 5 This is a schematic diagram of the structure of a PCBA reliability prediction device provided in an embodiment of the present application. Figure 5 As shown, the PCBA reliability prediction device 50 includes:

[0217] The acquisition module 501 is used to collect the temperature of the PCBA sample to obtain the measured temperature data;

[0218] A construction module 502 is used to construct a temperature field simulation sub-model, a mechanical field simulation sub-model, and an electric field simulation sub-model corresponding to the PCBA template according to the measured temperature data;

[0219] A coupling module 503 is used to couple the temperature field simulation sub-model, the mechanical field simulation sub-model, and the electric field simulation sub-model to obtain a multi-physics field simulation model;

[0220] A simulation module 504 is configured to simulate multiple failure modes using a multi-physics field simulation model to obtain multi-physics field distribution data corresponding to each failure mode;

[0221] The prediction module 505 is used to perform reliability prediction on the PCBA sample according to the multi-physics field distribution data corresponding to each failure mode to obtain reliability related information.

[0222] In a possible implementation, the acquisition module 501 is configured to:

[0223] The temperature of the PCBA sample is collected through a multi-modal sensor array to obtain raw temperature data; the multi-modal sensor array includes a thermocouple array, a distributed optical fiber temperature measurement unit, and an infrared thermal imaging unit;

[0224] The original temperature data is denoised to obtain the measured temperature data.

[0225] In a possible implementation, the construction module 502 is configured to:

[0226] For the PCBA sample, the heat conduction sub-model, convection heat dissipation sub-model and radiation heat transfer sub-model are constructed respectively;

[0227] The heat conduction sub-model, the convection heat dissipation sub-model and the radiation heat transfer sub-model are coupled to obtain a temperature field simulation sub-model; the temperature field simulation sub-model is expressed in the form of a nonlinear heat balance equation;

[0228] The Newton-Raphson algorithm is used to solve the nonlinear heat balance equation to obtain the temperature field distribution data under the current material parameters;

[0229] Determine the residual matrix based on the measured temperature data and temperature field distribution data of the PCBA sample;

[0230] Determine the residual root mean square error based on the residual matrix;

[0231] Determine whether the residual root mean square error is greater than the preset error value;

[0232] If the residual root mean square error is greater than the preset error value, the material parameters of the temperature field simulation sub-model are updated to obtain an optimized temperature field simulation sub-model;

[0233] Repeat the above iterative operation until the residual root mean square error is less than or equal to the preset error value.

[0234] In a possible implementation, the construction module 502 is further configured to:

[0235] Determine the temperature gradient between adjacent nodes in the simulation grid corresponding to the PCBA sample based on the temperature field distribution data;

[0236] If the temperature gradient is greater than the preset gradient value, the grid area where the adjacent nodes are located is meshed to obtain a new simulation grid.

[0237] In one possible implementation, the simulation module 504 is configured to:

[0238] For each failure mode, perform the following operations respectively, including:

[0239] Inject the failure mode into the temperature field simulation sub-model in a parameterized form;

[0240] Numerical solution is performed on the temperature field simulation sub-model to obtain temperature field distribution data;

[0241] Using the temperature field distribution data as input, the mechanical field simulation sub-model is driven to perform numerical solutions to obtain stress distribution data;

[0242] Using the temperature field distribution data as input, the electric field simulation sub-model is driven to perform numerical solution to obtain the current density distribution data.

[0243] In one possible implementation, the prediction module 505 is configured to:

[0244] Perform the following operations for each failure mode:

[0245] Determine the conductor resistance change data based on the temperature field distribution data after the fault mode injection and the temperature field distribution data before the fault mode injection;

[0246] Determine signal attenuation information based on wire resistance change data;

[0247] determining regional fracture information based on stress distribution data after the fault mode injection and stress distribution data before the fault mode injection;

[0248] Electromigration information is determined based on the current density distribution data after the fault mode injection and the current density distribution data before the fault mode injection.

[0249] In a possible implementation, the system further includes a detection module configured to:

[0250] Obtain the temperature field distribution data output by the temperature field simulation sub-model;

[0251] Determine whether there is an abnormality in the temperature field distribution data through a preset detection algorithm;

[0252] If there is an anomaly, an anomaly detection result is generated;

[0253] Match the anomaly detection results with the fault types recorded in the database to obtain the target fault type;

[0254] Feedback the anomaly detection results and target fault types to the manufacturing execution system.

[0255] In one possible implementation, the detection module is configured to perform at least one of the following:

[0256] Use the trained thermal behavior benchmark model to determine whether there are any anomalies in the temperature field distribution data; or,

[0257] Use the isolation forest-autoencoder hybrid algorithm to determine whether the temperature field distribution data is abnormal; or,

[0258] The trained temperature evolution prediction model is used to determine whether there are any anomalies in the temperature field distribution data.

[0259] In one possible implementation, the detection module is configured to:

[0260] Get the current working condition data of the PCBA sample;

[0261] Inputting the operating condition data into the thermal behavior benchmark model as input data, and outputting a predicted temperature range through the thermal behavior benchmark model;

[0262] Determine whether the temperature field distribution data is abnormal based on the predicted temperature range;

[0263] In a possible implementation, a training module is further included, configured to:

[0264] Obtain sample data and label data; the sample data includes historical operating condition data of the PCBA, and the label data includes the measured temperature range under the operating condition represented by the sample operating condition data;

[0265] The labeled data is used as the supervised data, and the initial thermal behavior benchmark model is iteratively trained according to the sample data to obtain a trained thermal behavior benchmark model.

[0266] In one possible implementation, the detection module is configured to:

[0267] The temperature field distribution data and the corresponding load parameters are used as input data into the temperature evolution prediction model, and the temperature field change information at each time point within the preset time period is predicted by the temperature evolution prediction model;

[0268] Determine whether there is a potential abnormality based on the temperature field change information.

[0269] In a possible implementation, a training module is further included, configured to:

[0270] The initial temperature evolution prediction model is iteratively trained by minimizing the loss function to obtain a trained temperature evolution prediction model; the loss function is determined based on the heat conduction equation and the data fitting error; the data fitting error is the mean square error between the predicted temperature and the sample temperature.

[0271] The description of the features in the embodiment corresponding to the PCBA reliability prediction device 50 can be found in the relevant description of the embodiment corresponding to the PCBA reliability prediction method, and will not be repeated here.

[0272] Figure 6 This is a schematic diagram of the structure of an electronic device provided in this application. Figure 1 The electronic device 40 shown. Figure 6 As shown, the electronic device 40 provided in this embodiment includes: a processor 401 and a memory 402. Optionally, the electronic device 40 further includes a communication component 403. The processor 401, the memory 402 and the communication component 403 are connected via a bus.

[0273] In a specific implementation process, the memory 402 is used to store a computer program. The processor 401 executes the computer program stored in the memory 402, so that the processor 401 executes the above-mentioned PCBA reliability prediction method embodiment.

[0274] The specific implementation process of the processor 401 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0275] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the application may be directly executed by a hardware processor or by a combination of hardware and software modules within the processor.

[0276] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage.

[0277] A bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be divided into address buses, data buses, and control buses.

[0278] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. The computer program is configured to execute the steps of any of the above-mentioned PCBA reliability prediction method embodiments when running.

[0279] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0280] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of any of the above-mentioned PCBA reliability prediction method embodiments are implemented.

[0281] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above-mentioned PCBA reliability prediction method embodiments are implemented.

[0282] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0283] The above describes in detail a PCBA reliability prediction solution provided by this application. This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from the principles of this application, and such improvements and modifications also fall within the scope of protection of the claims of this application.

Claims

1. A PCBA reliability prediction method, characterized in that: include: Collect the temperature of the printed circuit board assembly sample to obtain the measured temperature data; According to the measured temperature data, a temperature field simulation sub-model, a mechanical field simulation sub-model and an electric field simulation sub-model corresponding to the sample are respectively constructed; performing coupling processing on the temperature field simulation sub-model, the mechanical field simulation sub-model, and the electric field simulation sub-model to obtain a multi-physics field simulation model; Simulating multiple failure modes respectively through the multi-physics field simulation model to obtain multi-physics field distribution data corresponding to each failure mode; Performing reliability prediction on the sample according to the multi-physical field distribution data corresponding to each of the failure modes to obtain reliability-related information; The multi-physics field simulation model is used to simulate multiple failure modes to obtain multi-physics field distribution data corresponding to each failure mode, including: For each of the failure modes, the following operations are performed respectively, and the following operations include: Injecting the failure mode into the temperature field simulation sub-model in a parameterized form; Numerically solving the temperature field simulation sub-model to obtain temperature field distribution data; Using the temperature field distribution data as input, the mechanical field simulation sub-model is driven to perform numerical solution to obtain stress distribution data; Using the temperature field distribution data as input, the electric field simulation sub-model is driven to perform numerical solution to obtain current density distribution data; The performing reliability prediction on the sample according to the multi-physical field distribution data corresponding to each of the failure modes to obtain reliability-related information includes: For each of the failure modes, perform the following operations respectively, the following operations including: Determine the conductor resistance change data based on the temperature field distribution data after the fault mode injection and the temperature field distribution data before the fault mode injection; determining signal attenuation information based on the wire resistance change data; determining regional fracture information based on stress distribution data after the fault mode injection and stress distribution data before the fault mode injection; Electromigration information is determined based on the current density distribution data after the fault mode injection and the current density distribution data before the fault mode injection.

2. The method according to claim 1, characterized in that The temperature of the printed circuit board assembly sample is collected to obtain measured temperature data, including: The temperature of the sample is collected by a multimodal sensor array to obtain raw temperature data; the multimodal sensor array includes a thermocouple array, a distributed optical fiber temperature measurement unit and an infrared thermal imaging unit; The original temperature data is subjected to denoising processing to obtain the measured temperature data.

3. The method according to claim 1, characterized in that According to the measured temperature data, a temperature field simulation sub-model corresponding to the sample is constructed, including: Step 1: For the sample, construct a heat conduction sub-model, a convection heat dissipation sub-model and a radiation heat transfer sub-model respectively; Step 2: Coupling the heat conduction sub-model, the convection heat dissipation sub-model, and the radiation heat transfer sub-model to obtain the temperature field simulation sub-model; the temperature field simulation sub-model is expressed in the form of a nonlinear heat balance equation; Step 3: Solve the nonlinear heat balance equation using the Newton-Raphson algorithm to obtain temperature field distribution data under current material parameters; Step 4: determining a residual matrix based on the measured temperature data of the sample and the temperature field distribution data; Step 5: Determine the residual root mean square error according to the residual matrix; Step 6: Determine whether the residual root mean square error is greater than a preset error value; Step 7: If the residual root mean square error is greater than a preset error value, the material parameters of the temperature field simulation sub-model are updated to obtain an optimized temperature field simulation sub-model; Repeat steps 3 to 7 until the residual root mean square error is less than or equal to the preset error value.

4. The method according to claim 3, characterized in that Also includes: Determine the temperature gradient between adjacent nodes in the simulation grid corresponding to the sample according to the temperature field distribution data; If the temperature gradient is greater than a preset gradient value, the grid area where the adjacent nodes are located is meshed to obtain a new simulation grid.

5. The method according to claim 1, wherein Also includes: Acquire temperature field distribution data output by the temperature field simulation sub-model; Determine whether the temperature field distribution data is abnormal by using a preset detection algorithm; If there is an anomaly, an anomaly detection result is generated; Matching the anomaly detection result with the fault type recorded in the database to obtain the target fault type; The abnormality detection result and the target fault type are fed back to a manufacturing execution system.

6. The method according to claim 5, characterized in that Determining whether the temperature field distribution data is abnormal by using a preset detection algorithm includes at least one of the following: Using the trained thermal behavior benchmark model to determine whether the temperature field distribution data is abnormal; or, Using an isolation forest-autoencoder hybrid algorithm to determine whether the temperature field distribution data is abnormal; or, The trained temperature evolution prediction model is used to determine whether the temperature field distribution data is abnormal.

7. The method according to claim 6, characterized in that The step of using the trained thermal behavior benchmark model to determine whether the temperature distribution data is abnormal includes: Obtaining current working condition data of the sample; Inputting the operating condition data as input data into the thermal behavior benchmark model, and outputting a predicted temperature range through the thermal behavior benchmark model; Determine whether the temperature field distribution data is abnormal based on the predicted temperature range.

8. The method according to claim 7, characterized in that The training process of the thermal behavior benchmark model includes: Acquire sample data and label data; the sample data includes historical operating condition data of the printed circuit board assembly, and the label data includes a measured temperature range under the operating condition represented by the sample operating condition data; The label data is used as supervision data, and the initial thermal behavior benchmark model is iteratively trained according to the sample data to obtain the trained thermal behavior benchmark model.

9. The method according to claim 6, characterized in that The method of using the trained temperature evolution prediction model to determine whether the temperature field distribution data is abnormal includes: The temperature field distribution data and the corresponding load parameters are input into the temperature evolution prediction model as input data, and the temperature field change information at each time point within a preset time period in the future is rolling predicted by the temperature evolution prediction model; Determine whether there is a potential abnormality based on the temperature field change information.

10. The method according to claim 9, characterized in that The training process of the temperature evolution prediction model includes: The initial temperature evolution prediction model is iteratively trained by minimizing the loss function to obtain the trained temperature evolution prediction model; wherein the loss function is determined based on the heat conduction equation and the data fitting error; the data fitting error is the mean square error between the predicted temperature and the sample temperature.

11. An electronic device, characterized in that: include: memory for storing computer programs; A processor, configured to implement the steps of the PCBA reliability prediction method according to any one of claims 1 to 10 when executing the computer program.

12. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the PCBA reliability prediction method according to any one of claims 1 to 10 are implemented.

13. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the PCBA reliability prediction method according to any one of claims 1 to 10 are implemented.

Images