A method and system for establishing a digital twin of a circuit board and a terminal

Abstract

The application discloses a kind of circuit board digital twin's establishment method, system and terminal, by establishing digital twin life prediction model library and board card mechanism model library, and in digital twin life prediction model library and board card mechanism model library carry out prediction and analysis, accurate analysis circuit board life assessment and failure phenomenon prediction digital twin, with existing to be analyzed circuit board as analysis object, temperature humidity etc. after information collected by sensor is fused after the analysis of digital twin, comprehensive discrimination is carried out, the remaining service life of PCB and the failure phenomenon after failure are identified, and corresponding protection signal and alarm signal are generated, regular maintenance can be maintained or replaced in time when PCB is about to fail, improve the availability of electronic terminal. While being able to carry out high-precision life prediction and failure prediction, the digital twin built can also realize rapid and convenient modeling and real-time calculation with low computing power requirements.

Description

Technical Field

[0001] This invention relates to the field of printed circuit board life assessment technology, specifically to a method, system, and terminal for establishing a digital twin of a circuit board. Background Technology

[0002] With the continuous development of science and technology, the application rate of electronic technology in terminal products continues to increase. Due to the significant advantages of electronic terminal devices, large-scale application of electronic products has become an unstoppable trend. Printed Circuit Boards (PCBs) are known as the "mother of electronic products." As an indispensable component of electronic terminal devices, the reliability of PCBs directly affects whether a terminal device can successfully achieve its intended goals.

[0003] As electronic products become increasingly complex, the deformation, heat transfer, and stress of PCBs during actual operation will affect the reliability of the entire circuit board. This results in multiple failure modes during application. As the carrier of most electronic products, the reliability of PCBs during operation has always been a key focus in the manufacturing industry. How to conduct an overall life assessment of PCBs is a major challenge at present.

[0004] PCB failures can cause significant damage, ranging from minor electronic device malfunctions to life-threatening situations. Statistics from the U.S. Air Force show that over half of the failure modes in military aircraft electronics are related to their operating temperature environment. Furthermore, this research indicates two typical failure modes in electronic devices operating under temperature conditions: the first type is functional failure of components due to prolonged temperature loads (component-related failure modes); the second type is solder joint fatigue failure caused by repeated temperature cycles (solder joint fatigue failure modes).

[0005] Traditional PCB analysis mainly focuses on reliability or lifespan assessment, and the analysis methods are mainly divided into two types:

[0006] One approach involves establishing a lumped model that analyzes individual units on the PCB, such as the characteristics of a single electronic component, connector, or solder joint, and then modeling by modifying empirical formulas. However, this method can only analyze individual working units on the PCB using a lumped parameter model; it cannot analyze the overall operating state of the PCB or its behavior under fault conditions.

[0007] Secondly, three-dimensional analysis methods such as finite element method are used to analyze the entire circuit board. This analysis method can analyze the entire board or a specific part of the board. This method can predict both fault phenomena and board lifespan. This analysis method has very high accuracy. The biggest drawback is that even a tiny prediction requires a lot of computing resources.

[0008] Therefore, existing prediction methods struggle to achieve a balance between modeling speed, computational accuracy, and computational resource consumption. Summary of the Invention

[0009] In order to overcome the shortcomings of the existing technology, the purpose of this invention is to provide a method, system and terminal for establishing a digital twin of a circuit board, so as to solve the problem of large computing resources when performing overall analysis of a circuit board in the existing technology.

[0010] This invention is achieved through the following technical solution:

[0011] A method for creating a digital twin of a circuit board includes the following steps:

[0012] Step 1: Establish a digital twin lifespan prediction model library and a board mechanism model library. In the digital twin lifespan prediction model library, establish the board lifespan prediction model through drag-and-drop modeling. In the board mechanism model library, establish the board mechanism model through drag-and-drop modeling.

[0013] Step 2: Read the historical analysis data of the board and calculate the lifespan of the board and vulnerable components in real time through the lifespan prediction model of the board. When there is fault information in the lifespan of the board and vulnerable components, analyze the phenomenon of potential fault units failing through the mechanism model of the board. Otherwise, obtain the analysis data of the board and the expected lifespan of the board.

[0014] Step 3: In the mechanism model of the board, determine whether the potential fault information will endanger the operation of the entire system. If it will endanger the operation of the entire system, output the expected fault phenomenon of the board with an alarm. Otherwise, output the expected usage plan to the mechanism model of the board to complete the establishment of the digital twin of the circuit board.

[0015] Preferably, in step 1, the digital twin lifetime prediction model library includes electronic components, contact devices, and PCB boards, wherein the electronic components include, but are not limited to, semiconductor integrated circuits such as resistors, capacitors, inductors, and light-emitting diodes.

[0016] Contact devices include, but are not limited to, switches, mechanical relays, and connectors;

[0017] PCBs include, but are not limited to, FR-4, FPC, aluminum substrate, thermoelectrically separated copper substrate, Rogers high-frequency board, PTFE Teflon high-frequency board and circuit board;

[0018] The board mechanism model library includes, but is not limited to, electronic devices and connection devices, among which electronic devices include resistors, capacitors, inductors and semiconductor devices;

[0019] Connection devices include connectors, solder joints, and copper plating on PCBs.

[0020] Preferably, in step 1, the digital twin lifetime prediction model library includes a model library selection section and a work area section. The model is dragged from the model library to the work area section to complete the model selection. The model is selected in the work area section and the model parameter setting interface is entered to set the model parameters. The set model can be copied and pasted for reuse of models with the same parameters.

[0021] Preferably, in step 1, the board mechanism model library includes a model library area and a working area. After dragging the model from the model library area to the working area, the parameters are set, and the devices located in the working area are connected according to the same connection relationship as the real board to complete the modeling. During the modeling process, the pins of the two electronic components are connected by copper plating.

[0022] Preferably, in step 1, the digital twin lifespan prediction model library processes temperature, humidity, and pressure to establish independent analysis models for different types of objects and sets parameters individually, wherein the independent analysis models have historical data insertion functions.

[0023] Preferably, in step 1, the board life prediction model sets multiple confidence levels for the board life analysis process and outputs the predicted life under different confidence levels. The expected life curve is compared with the confidence level to obtain the expected life of the board. The confidence level is determined by the quality of the components used in the board itself and the degree of trust in the board.

[0024] Preferably, in step 2, the board's lifespan prediction model predicts the board's lifespan and vulnerable components. The specific lifespan prediction process is as follows:

[0025] The model checks whether the board has historical data. If there is historical data, the model needs to set the initial lifespan data to the historical data during calculation; otherwise, the model sets the lifespan data to 100 before calculation.

[0026] After the historical data is analyzed, the board's lifespan prediction model incorporates the data read by the sensors into the algorithm, and calculates the board's wear based on the temperature, humidity, and pressure read by the sensors.

[0027] Preferably, in step 2, when there is fault information regarding the lifespan of the board and vulnerable components, the fault phenomenon of the shortest lifespan element is calculated through the board's mechanism model.

[0028] A system for creating a digital twin of a circuit board includes:

[0029] The model building module is used to build a digital twin lifespan prediction model library and a board mechanism model library. Within the digital twin lifespan prediction model library, the lifespan prediction model of the board is built through drag-and-drop modeling. Within the board mechanism model library, the mechanism model of the board is built through drag-and-drop modeling.

[0030] The lifespan prediction module is used to read historical analysis data of the board and calculate the lifespan of the board and vulnerable components in real time through the lifespan prediction model of the board. When there is fault information in the lifespan of the board and vulnerable components, the mechanism model of the board is used to analyze the phenomenon of potential fault units failing. Otherwise, the analysis data of the board and the expected lifespan of the board are obtained.

[0031] The fault analysis module is used to determine whether potential fault information in the board's mechanism model will jeopardize the operation of the entire system. If it does, an alarm is outputting the expected fault phenomenon of the board; otherwise, the expected usage plan is output to the board's mechanism model to complete the establishment of the digital twin of the circuit board.

[0032] A mobile terminal, wherein the computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for establishing a circuit board digital twin as described above.

[0033] Compared with the prior art, the present invention has the following beneficial technical effects:

[0034] This invention provides a method for establishing a digital twin of a circuit board. By establishing a digital twin lifespan prediction model library and a circuit board mechanism model library, and performing predictions and analyses within these libraries, an accurate digital twin for circuit board lifespan assessment and fault phenomenon prediction is generated. Using an existing circuit board as the analysis object, information such as temperature and humidity collected by sensors is analyzed by the digital twin and then fused for comprehensive judgment. This identifies the remaining lifespan of the PCB and the fault phenomena that will occur after a failure, generating corresponding protection and alarm signals. During regular maintenance, PCBs that are about to fail can be maintained or replaced in a timely manner, improving the availability of electronic terminals. While enabling high-precision lifespan and fault prediction, the constructed digital twin also allows for rapid and convenient modeling and real-time calculations with low computational requirements. Attached Figure Description

[0035] Figure 1This is a flowchart of the method for establishing a digital twin of a circuit board in this invention;

[0036] Figure 2 This is a schematic diagram of the architecture of the digital twin for circuit board life assessment in an embodiment of the present invention;

[0037] Figure 3 This is the ambient temperature curve of the PCB during operation in an embodiment of the present invention;

[0038] Figure 4 This refers to the PCB lifetime prediction results in the embodiments of the present invention;

[0039] Figure 5 This is the PCB fault phenomenon prediction result in the embodiment of the present invention.

[0040] In the diagram: 1-PCB unit to be analyzed; 2-First communication line; 3-Sensor; 4-Second communication line; 5-Lifetime prediction computer; 6-PCB board lifetime prediction toolbox; 7-PCB fault phenomenon prediction toolbox; 8-Board lifetime prediction model library; 9-Board lifetime analysis and solution tools; 10-PCB working principle model library; 11-PCB fault introduction tool; 12-Electronic component lifetime prediction model; 13-Solder joint lifetime prediction model; 14-Contact device lifetime prediction model; 15-PCB substrate lifetime prediction model; 16-Historical lifetime analysis. Analysis tools; 17-Real-time life analysis tool; 18-Electronic component working mechanism model; 19-Solder joint working mechanism model; 20-Contact device working mechanism model; 21-PCB substrate working mechanism model; 22-PCB short circuit fault introduction module; 23-PCB open circuit fault introduction module; 24-First electronic component unit; 25-First surface mount via unit; 26-First connector unit; 27-First PCB board; 28-Second electronic component unit; 29-Second surface mount via unit; 30-Second connector unit; 31-Second PCB board. Detailed Implementation

[0041] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0042] The present invention will now be described in further detail with reference to the accompanying drawings:

[0043] The purpose of this invention is to provide a method, system, and terminal for establishing a digital twin of a circuit board, so as to solve the problem of large computing resources when performing overall analysis of a circuit board in the prior art.

[0044] Specifically, according to Figure 1 As shown, the method for creating a digital twin of the circuit board includes the following steps:

[0045] Step 1: Establish a digital twin lifespan prediction model library and a board mechanism model library. In the digital twin lifespan prediction model library, establish the board lifespan prediction model through drag-and-drop modeling. In the board mechanism model library, establish the board mechanism model through drag-and-drop modeling.

[0046] Specifically, the model library was established to enable the reuse of existing models when modeling different circuit boards. The main feature of the model library is that it supports drag-and-drop zero-code modeling. When modeling begins, you only need to drag the model you need from the model library to the workspace, and then connect the models with lines to complete the modeling. This modeling method can greatly improve modeling efficiency.

[0047] The life prediction model library used in PCB boards contains a variety of life calculation methods, and the algorithm is automatically or manually selected for different operating conditions of the PCB (temperature, humidity, etc.).

[0048] Specifically, the digital twin lifetime prediction model library includes electronic components, contact devices, and PCB boards. Among them, electronic components include, but are not limited to, semiconductor integrated circuits such as resistors, capacitors, inductors, and light-emitting diodes.

[0049] Contact devices include, but are not limited to, switches, mechanical relays, and connectors;

[0050] PCBs include, but are not limited to, FR-4, FPC, aluminum substrate, thermoelectrically separated copper substrate, Rogers high-frequency board, PTFE Teflon high-frequency board and circuit board;

[0051] The board mechanism model library includes, but is not limited to, electronic devices and connection devices, among which electronic devices include resistors, capacitors, inductors and semiconductor devices;

[0052] Connection devices include connectors, solder joints, and copper plating on PCBs.

[0053] Specifically, the digital twin lifetime prediction model library includes a model library selection section and a work area section. Models are selected by dragging and dropping them from the model library to the work area section. In the work area section, the model is selected and the model parameter settings interface is entered to set the model parameters. The set model can be copied and pasted for reuse of models with the same parameters.

[0054] Models dragged from the model library to the analysis area can have all their parameters easily adjusted to meet the model's lifetime prediction needs. The models in the library include the ability to import historical data and record real-time calculated data. After dragging a model to the workspace, the model automatically accumulates historical data. The models in the library also include the ability to process data from sensors.

[0055] Specifically, the board mechanism model library includes a model library area and a working area. After dragging a model from the model library area to the working area, parameters are set, and the components in the working area are connected according to the same connection relationships as the actual board to complete the modeling. During the modeling process, the pins of two electronic components are connected through copper plating. The established mechanism model should be able to reflect all the characteristics of the board. The dragged model can automatically form a digital twin lifetime prediction network with the models already dragged to the working area.

[0056] Specifically, the digital twin life prediction model library processes temperature, humidity, and pressure to establish independent analysis models for different types of objects and sets parameters individually. These independent analysis models have the function of inserting historical data.

[0057] Specifically, the board life prediction model sets multiple confidence levels for the board life analysis process and outputs the predicted life at different confidence levels. The expected life curve is compared with the confidence level to obtain the expected life of the board. The confidence level is determined by the quality of the components used in the board and the degree of trust in the board.

[0058] Step 2: Read the historical analysis data of the board and calculate the lifespan of the board and vulnerable components in real time through the lifespan prediction model of the board. When there is fault information in the lifespan of the board and vulnerable components, analyze the phenomenon of potential fault units failing through the mechanism model of the board. Otherwise, obtain the analysis data of the board and the expected lifespan of the board.

[0059] Specifically, the circuit board life prediction model predicts the lifespan of the circuit board and vulnerable components. The specific life prediction process is as follows:

[0060] The model checks whether the board has historical data. If there is historical data, the model needs to set the initial lifespan data to the historical data during calculation; otherwise, the model sets the lifespan data to 100 before calculation.

[0061] After the historical data is analyzed, the board's lifespan prediction model incorporates the data read by the sensors into the algorithm, and calculates the board's wear based on the temperature, humidity, and pressure read by the sensors.

[0062] In a typical analysis environment, a digital twin lifetime prediction model should at least take into account the effect of temperature. This invention considers various effects, with the most typical example being the effect of temperature on PCB lifetime under temperature cyclic loading.

[0063] Step 3: In the mechanism model of the board, determine whether the potential fault information will endanger the operation of the entire system. If it will endanger the operation of the entire system, output the expected fault phenomenon of the board with an alarm. Otherwise, output the expected usage plan to the mechanism model of the board to complete the establishment of the digital twin of the circuit board.

[0064] In this invention, various types of sensors can use sensors already installed on the board for analysis using shared data, or sensors can be set up individually on the board or in the rack.

[0065] The established model should be flexible enough to allow all configurable parameters to be made available, so that operators can customize settings for different types of devices.

[0066] The digital twin lifetime prediction model should have multiple built-in calculation methods, allowing operators to easily choose between calculation accuracy and speed. A high-speed, low-accuracy mode is used to quickly view lifetime trends, while a high-accuracy, low-speed mode is used to view accurate lifetime curves.

[0067] The lifetime analysis process should allow setting multiple confidence levels and outputting predicted lifetimes at different confidence levels. Before performing lifetime analysis, the digital twin should be able to load historical data from the board; simultaneously, the data from the current analysis should be displayed and stored in real time.

[0068] The circuit board mechanism model is established based on the working principles of various components and interconnection devices, and can truly reflect the input and output effects of the circuit board.

[0069] The fault introduction model is directly linked to the board's lifespan prediction model. When the lifespan prediction identifies faulty components, the fault introduction model can automatically match faults that may occur during that period. Typical faults include, but are not limited to, short circuits and open circuits at both ends of the device caused by various reasons.

[0070] Example 1

[0071] This embodiment provides an architecture for establishing a digital twin of a circuit board, based on... Figure 2As shown, it specifically includes a PCB unit 1 to be analyzed, a sensor 3, and a life prediction computer 5. The PCB unit 1 to be analyzed is connected to one end of the sensor 3 through a first communication line 2, and the other end of the sensor 3 is connected to the life prediction computer 5 through a second communication line 5. The life prediction computer 5 includes a PCB board life prediction toolbox 6 and a PCB failure phenomenon prediction toolbox 7.

[0072] The PCB life prediction toolbox 6 includes a board life prediction model library 8 and a board life analysis and solution tool 9. The board life prediction model library 8 includes an electronic component life prediction model 12, a solder joint life prediction model 13, a contact device life prediction model 14, and a PCB substrate life prediction model 15. The electronic component life prediction model 12 includes a first electronic component unit 24 (resistor, capacitor, inductor, chip, semiconductor, LED, etc.); the solder joint life prediction model 13 includes a first surface mount via unit 25 (surface mount, via, etc.); the contact device life prediction model 14 includes a first connection device unit 26 (connector, switch, etc.); and the PCB substrate life prediction model 15 includes a first PCB board 27 (FR4, aluminum substrate, etc.). The board life analysis and solution tool 9 includes a historical life analysis tool 16 and a real-time life analysis tool 17.

[0073] The PCB fault prediction toolbox 7 includes a PCB working principle model library 10 and a PCB fault introduction tool 11. The PCB working principle model library 10 includes an electronic component working mechanism model 18, a solder joint working mechanism model 19, a contact device working mechanism model 20, and a PCB substrate working mechanism model 21. Among them, the electronic component working mechanism model 18 includes a second electronic component unit 28 (resistor, capacitor, inductor, chip, semiconductor, LED, etc.); the solder joint working mechanism model 19 includes a second surface mount via unit 29 (surface mount, via, etc.); the contact device working mechanism model 20 includes a second connection device unit 30 (connector, switch, etc.); the PCB substrate working mechanism model 21 includes a second PCB board (FR4, aluminum substrate, etc.); and the PCB fault introduction tool 11 includes a PCB open circuit fault introduction module 22 and a PCB open circuit fault introduction module 23.

[0074] Example 2

[0075] This embodiment is used to calculate the lifespan and failure phenomena of a circuit board operating under cyclic temperature. The main function of the circuit board is to perform signal isolation, inputting a 0-20mA analog signal and outputting an isolated 0-20mA analog signal.

[0076] S1. First, establish a life prediction model library for the circuit board. This specific implementation takes the life prediction model of components and solder joints as an example:

[0077] S11, Component Lifetime Distribution: The lifespan of components within the same batch is not entirely consistent; the randomness of lifespans across different samples needs to be described using a lifetime distribution. Since the lifespan of most electronic components is considered to follow an exponential distribution, the industry typically uses this distribution to predict component lifespan distribution. The most prominent characteristic of equipment with an exponentially distributed lifespan is that its failure rate remains constant across different stages of its lifespan. The distribution function of the exponential distribution is as follows:

[0078] f(t)=1-exp(-λt),t≥0 (1)

[0079] Here, λ is the only parameter in the distribution function, and its statistical meaning is the failure rate of electronic components.

[0080] Taking the derivative of the distribution function of the exponential distribution yields the probability density function of the exponential distribution:

[0081] f(t)=λexp(-λt),t≥0 (2)

[0082] Accelerated failure models for electronic components: Accelerated failure models are used to describe the relationship between a product's lifespan and reliability metrics and the stress environment. The Arrhenius model is commonly used for calculating the accelerated lifespan of electronic components.

[0083] The Arrhenius acceleration model was proposed by Arrhenius in 1880 based on a large amount of data obtained from a series of chemical reactions. According to the Arrhenius model, the relationship between a product's lifespan characteristic and the temperature environment in which the product is located can be expressed by the following formula:

[0084]

[0085] In the formula, L represents a certain lifespan characteristic of the product, which can be obtained through statistics; A is a constant related to the product, which can be obtained through experiments; E a The activation energy required to excite potential defects in the product; k is the Boltzmann constant; T is the absolute temperature, i.e., the temperature in the working environment of the board.

[0086] For component failure modes, it is generally assumed that the failure rate remains constant throughout its lifespan, and the failure time follows an exponential distribution. Based on the exponential failure distribution function shown in formula (4), the reliability function for component failure modes can be obtained as follows:

[0087] R(t)=1-F(t)=exp(-λt) (4)

[0088] Based on the reliability function shown in the above formula and the exponential probability density function shown in formula (4), the failure rate function of component failure modes can be obtained as follows:

[0089]

[0090] Using the Arrhenius model shown in formula (5), the relationship between the mean failure time θ(T) of component failure modes and ambient temperature can be described as follows:

[0091]

[0092] For products whose lifespan follows an exponential distribution, the failure rate under constant temperature and load conditions is equal to the reciprocal of the mean time to failure, i.e.

[0093]

[0094] In the formula, Furthermore, the reliability function of component failure modes under constant temperature load conditions can be obtained as follows:

[0095]

[0096] The temperature in a temperature cycle changes periodically over time, that is...

[0097] T=g(t) (9)

[0098] Therefore, even if the failure time of a product's component failure modes follows an exponential distribution, its failure rate in the working environment is not constant, but rather varies continuously with the ambient temperature.

[0099]

[0100] Based on the above equation, substituting (10) into equation (2), we can obtain the expression for the failure probability density function of component failure modes under temperature cyclic loading conditions as follows:

[0101]

[0102] Based on the failure probability density function in the above equation, the reliability function of component failure modes under temperature cyclic loading conditions can be further obtained as follows:

[0103]

[0104] In the formula: t is the operating time of the circuit board under cyclically changing temperature; A is a constant related to the product; E aThe activation energy required to excite potential defects in a product; k is the Boltzmann constant; g(t) is a function of temperature that changes periodically over time, or real-time temperature data can be used.

[0105] S12, Solder Joint Reliability Model: Currently, the most widely used high and low temperature alternating acceleration model is the Norris-Landzberg model. This model is suitable for describing the fatigue life of electromechanical products under high and low temperature alternating conditions. The expression of the Norris-Landzberg model is:

[0106]

[0107] In the formula, t TM This refers to the high-temperature holding time. The improved Norris-Landzberg model can effectively describe the effects of creep and stress relaxation on the fatigue life of weld joints, and is more suitable for describing the mathematical model of weld joint fatigue life under different temperature cyclic loading conditions.

[0108] The failure rate of some electronic components cannot remain constant throughout their lifespan, making it impossible to describe their lifespan using an exponential distribution. The Weibull distribution, with its multiple distribution parameters (two-parameter Weibull distribution, three-parameter Weibull distribution), can better describe the randomness of product lifespans where failure rates cannot remain constant.

[0109] The distribution function of the two-parameter Weibull distribution is:

[0110]

[0111] The parameters of the two-parameter Weibull distribution include the scale parameter η and the shape parameter β. The probability density function of the two-parameter Weibull distribution is:

[0112]

[0113] The three-parameter Weibull distribution adds a location parameter γ to the two-parameter Weibull distribution. The distribution function and probability density function of the three-parameter Weibull distribution are as follows:

[0114]

[0115]

[0116] For solder joint fatigue failure modes, a two-parameter Weibull distribution is used to describe the distribution of failure time. Based on the failure distribution function of the two-parameter Weibull distribution shown in formula (14), the reliability function for solder joint fatigue failure modes can be obtained as follows:

[0117]

[0118] Further, based on the Weibull probability density function shown in formula (15), the failure rate function for fatigue-related failure modes of weld joints can be obtained as follows:

[0119]

[0120] The fatigue life of solder joint fatigue failure modes under different temperature cycles is described using the Norris-Landzberg model shown in Equation (13). The failure time of solder joint fatigue failure modes is equal to the product of the temperature cycle period and the fatigue life of the failure mode, i.e.

[0121]

[0122] Based on the above formula, the relationship between the characteristic lifetime η of solder joint fatigue failure modes and the temperature cycling parameters can be described as follows:

[0123]

[0124] In the formula, c = lnδ; d = m; e = -n; g = Ea / k.

[0125] In accelerated testing statistical analysis, it is generally assumed that the shape parameter of the Weibull distribution does not change with the accelerated testing stress. Substituting equation (21) into equations (18) and (19) respectively, the reliability function of the weld fatigue failure mode under different temperature cyclic loading conditions can be obtained as follows:

[0126]

[0127] In the formula, t is the operating time of the board to be analyzed; f is the frequency of temperature cycling; t TM ΔT represents the duration of the highest temperature range for the circuit board during a temperature cycle; e is the natural constant, taken as 2.71828; ΔT is the temperature cycle range, i.e., the difference between the highest and lowest temperatures of a circuit board during one cycle; g = E a / k, E a The activation energy required to trigger a potential defect in a product, also known as the activation energy, is the energy required for a component on the circuit board to malfunction; k is the Boltzmann constant.

[0128] The failure rate function for fatigue-related failure modes of solder joints under different temperature cyclic loading conditions is as follows:

[0129]

[0130] S13, the above process constructs the life prediction model of components and solder joints. After modeling various types of objects, a life prediction model library for circuit boards can be formed.

[0131] S2, by combining models in the model library according to the number and construction method of real boards, establishes a life prediction model for a specific target board.

[0132] S3, Read historical analysis data from the board.

[0133] Before continuing analysis on a board that has already been analyzed, the digital twin needs to read the board's previous analysis data and then continue the analysis based on that data. If there is no historical analysis data, the digital twin will treat the board as a completely new board for analysis.

[0134] S4, the lifespan prediction model calculates the board's lifespan and vulnerable components in real time.

[0135] The life prediction model is based on a set of equations that combine the life prediction models of components and solder joints on the circuit board to calculate the life of the circuit board under cyclic temperature and the most vulnerable components on the circuit board.

[0136] S5 records the analysis data of the board in real time.

[0137] The digital twin records the board's operating data in real time while it is in operation, which facilitates continued analysis when the board is used again after it has stopped being used.

[0138] S6, Comparing lifetime with confidence level

[0139] The expected lifespan of the board is determined by comparing the expected lifespan curve with the confidence level. The confidence level setting is determined by factors such as the quality of the components used in the board and the degree of trust in the board.

[0140] S7, Expected lifespan of the real-time output board

[0141] In this embodiment, the digital twin outputs the expected lifespan of the board in real time during the calculation process.

[0142] S8, Establishing a board mechanism model library

[0143] Mechanism model of components considering temperature:

[0144] In the resistor model of the circuit board, if the value of TCE is defined, then the formula for calculating the resistance value is:

[0145]

[0146] If the value of TCE is not defined, then the formula for calculating the resistance is:

[0147]

[0148] The formula for calculating the capacitance value considering temperature is as follows:

[0149] value=C×(1+VC1×V+VC2×V 2 )×(1+TC1×(TT nom )+TC2×(TT nom ) 2 (26)

[0150] Where: TCE: Temperature Coefficient; T: Analysis Temperature, i.e., the operating temperature of the component; Tnom: Nominal Temperature, i.e., the standard operating temperature of the component; VC: Voltage between control nodes, i.e., the voltage on the PCB of the component.

[0151] Voltage across the components; TC1: linear temperature coefficient; TC2: quadratic temperature coefficient;

[0152] S9 uses the board mechanism model library to build the board mechanism model through drag-and-drop modeling.

[0153] The establishment of the board mechanism model mainly uses the basic laws of circuits. In this embodiment, Kirchhoff's first law (KCL) and Kirchhoff's second law (KVL) are used to calculate the board mechanism model composed of various component models.

[0154] Kirchhoff's First Law (KCL):

[0155] Expressed as an equation, for any node in the circuit, the following holds true:

[0156]

[0157] Where is the current entering or leaving this node, and is the current flowing through the kth branch connected to this node, which can be a real number or a complex number.

[0158] Kirchhoff's Second Law (KVL):

[0159] Expressed as an equation, for any closed loop in a circuit,

[0160]

[0161] Where m is the number of elements in this closed loop, v k It is the voltage across the component, which can be a real number or a complex number.

[0162] S10, Analyze the phenomena of potential faulty units.

[0163] The entire circuit board's mechanism model is built using a circuit board mechanism model library. The mechanism model calculates the fault phenomena of the circuit board after a fault occurs at a certain location.

[0164] Does S11 jeopardize the operation of the entire system?

[0165] Digital twins can determine whether a fault will affect the operation of the entire board or even the entire electronic device.

[0166] S12, providing the expected usage plan.

[0167] If the digital twin determines that the predicted fault will not affect the operation of the entire system, then the digital twin analyzes and evaluates the expected lifespan of the board and provides a prompt on whether the board can continue to be used.

[0168] S13, alarm and output the expected fault symptoms of the board.

[0169] If a predicted failure would severely impact system operation, the digital twin would issue a warning, reminding the user of the board's expected lifespan and the serious failure that is about to occur.

[0170] The operating temperature of the digital twin in this invention:

[0171] In this specific real-time method, a digital twin of the analog input / output isolation board was built, and the working effect of the digital twin under a 24-hour cyclic temperature was analyzed. The specific temperature curve is shown below. Figure 3 As shown.

[0172] Results of the digital twin project:

[0173] In this specific embodiment, a digital twin constructed through the above process is used to predict the lifespan of the target board and the failure phenomena after a failure. Regarding lifespan prediction, the lifespan prediction model predicts the lifespan of the target board at different confidence levels, and the specific results are as follows: Figure 4 As shown.

[0174] In the above results, fault point 1 is the location most prone to failure calculated by the constructed digital twin, and fault point 2 is the location second most prone to failure calculated by the digital twin. Confidence level refers to the degree of reliance on that location. For example, a confidence level of 0.7 indicates a 70% probability of the board failing, while a confidence level of 0.95 indicates a 95% probability of failure. Therefore, the higher the confidence level, the more likely the location is to fail, and correspondingly, the shorter the lifespan of the board.

[0175] The selection of confidence level is affected by the manufacturing quality of components on the PCB board and the usage environment. It is a result obtained through statistics. In order to introduce the performance of digital twin lifetime prediction under different confidence levels, this embodiment sets multiple confidence level values.

[0176] After completing lifespan prediction, the digital twin predicts board failure phenomena. Based on the lifespan prediction, the digital twin's failure phenomenon prediction function begins to function at the weak point affecting the board's lifespan, i.e., the first failure analysis point. This specific embodiment analyzes the overall board response under different failure severity levels after a failure occurs at the first failure analysis point, such as... Figure 5 As shown.

[0177] In the digital twin's prediction of board fault phenomena, the first curve in the upper left corner represents the board's signal input, and the remaining curves represent the board's signal output. It can be seen that when the fault level is low (0.0-0.3), the board's output and input signals are consistent. This is because the board has a certain fault tolerance capability; for example, the power supply chip has a power supply range within which it can operate. When the fault level exceeds 0.4, the board's output signal becomes distorted due to the fault, and the more severe the fault level, the more severe the output distortion.

[0178] Calculation time:

[0179] The computation speed of the board digital twin depends on the computer's performance and the size of the simulation model. On a computer with an i7 8700 processor, it takes 2.09 seconds of simulation time to run the model for 5 seconds. Since the model preprocessing and compilation take a lot of time, the overall simulation time percentage will be lower as the model running time increases. For example, a 500-second model simulation only requires 9.5 seconds of computation, as shown in Table 1.

[0180] 5 seconds 2.11 seconds 10 seconds 2.34 seconds 50 seconds 3.22 seconds 100 seconds 4.08 seconds 200 seconds 5.73 seconds 500 seconds 9.5 seconds

[0181] Table 1 is a statistical table of the actual calculation time consumed under different model simulation times.

[0182] Calculation accuracy:

[0183] In the process of establishing the mechanism model of the board digital twin, almost all components of the real board schematic are fully referenced and reflected. The maximum error between the board's input and output is only 0.00110258961178mA.

[0184] In summary, this invention provides a method for establishing a digital twin of a circuit board. By establishing a digital twin lifespan prediction model library and a circuit board mechanism model library, and performing predictions and analyses within these libraries, an accurate digital twin for circuit board lifespan assessment and fault phenomenon prediction is generated. Using an existing circuit board as the analysis object, information such as temperature and humidity collected by sensors is analyzed by the digital twin and then fused for comprehensive judgment. This identifies the remaining lifespan of the PCB and the fault phenomena that will occur after a failure, generating corresponding protection and alarm signals. During regular maintenance, PCBs that are about to fail can be maintained or replaced in a timely manner, improving the availability of electronic terminals. While enabling high-precision lifespan and fault prediction, the constructed digital twin also allows for rapid and convenient modeling and real-time calculations with low computational requirements.

[0185] This invention achieves lifespan prediction and fault phenomenon prediction for the entire circuit board through lifespan prediction and fault prediction models. Simultaneously, this invention establishes a series of model libraries, enabling rapid modeling of the analyzed object through model library technology. The use of multiple mathematical models significantly improves modeling and computational efficiency, meeting the requirements of real-time solution, real-time analysis, and real-time early warning in modern electronic terminals.

[0186] The present invention also provides a system for creating a digital twin of a circuit board, comprising:

[0187] The model building module is used to build a digital twin lifespan prediction model library and a board mechanism model library. Within the digital twin lifespan prediction model library, the lifespan prediction model of the board is built through drag-and-drop modeling. Within the board mechanism model library, the mechanism model of the board is built through drag-and-drop modeling.

[0188] The lifespan prediction module is used to read historical analysis data of the board and calculate the lifespan of the board and vulnerable components in real time through the lifespan prediction model of the board. When there is fault information in the lifespan of the board and vulnerable components, the mechanism model of the board is used to analyze the phenomenon of potential fault units failing. Otherwise, the analysis data of the board and the expected lifespan of the board are obtained.

[0189] The fault analysis module is used to determine whether potential fault information in the board's mechanism model will jeopardize the operation of the entire system. If it does, an alarm is outputting the expected fault phenomenon of the board; otherwise, the expected usage plan is output to the board's mechanism model to complete the establishment of the digital twin of the circuit board.

[0190] The present invention also provides a terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, such as a program for creating a digital twin of a circuit board.

[0191] When the processor executes the computer program, it implements the steps of the method for establishing the digital twin of the circuit board described above, for example, including the following steps:

[0192] Step 1: Establish a digital twin lifespan prediction model library and a board mechanism model library. In the digital twin lifespan prediction model library, establish the board lifespan prediction model through drag-and-drop modeling. In the board mechanism model library, establish the board mechanism model through drag-and-drop modeling.

[0193] Step 2: Read the historical analysis data of the board and calculate the lifespan of the board and vulnerable components in real time through the lifespan prediction model of the board. When there is fault information in the lifespan of the board and vulnerable components, analyze the phenomenon of potential fault units failing through the mechanism model of the board. Otherwise, obtain the analysis data of the board and the expected lifespan of the board.

[0194] Step 3: In the mechanism model of the board, determine whether the potential fault information will endanger the operation of the entire system. If it will endanger the operation of the entire system, output the expected fault phenomenon of the board with an alarm. Otherwise, output the expected usage plan to the mechanism model of the board to complete the establishment of the digital twin of the circuit board.

[0195] Alternatively, when the processor executes the computer program, it implements the functions of each module in the above system, such as: a model building module, used to build a digital twin life prediction model library and a board mechanism model library, and to build a life prediction model of the board through drag-and-drop modeling in the digital twin life prediction model library, and to build a mechanism model of the board through drag-and-drop modeling in the board mechanism model library.

[0196] The lifespan prediction module is used to read historical analysis data of the board and calculate the lifespan of the board and vulnerable components in real time through the lifespan prediction model of the board. When there is fault information in the lifespan of the board and vulnerable components, the mechanism model of the board is used to analyze the phenomenon of potential fault units failing. Otherwise, the analysis data of the board and the expected lifespan of the board are obtained.

[0197] The fault analysis module is used to determine whether potential fault information in the board's mechanism model will jeopardize the operation of the entire system. If it does, an alarm is outputting the expected fault phenomenon of the board; otherwise, the expected usage plan is output to the board's mechanism model to complete the establishment of the digital twin of the circuit board.

[0198] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the mobile terminal.

[0199] The mobile terminal can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. The mobile terminal may include, but is not limited to, a processor and a memory.

[0200] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the mobile terminal, connecting various parts of the mobile terminal via various interfaces and lines.

[0201] The memory can be used to store the computer program and / or module. The processor implements various functions of the mobile terminal by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0202] The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function (such as sound playback, image playback, etc.); the data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). Furthermore, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital cards (SD cards), flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0203] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method of establishing a digital twin of a circuit board, characterized by, Includes the following steps: Step 1: Establish a digital twin lifespan prediction model library and a board mechanism model library. In the digital twin lifespan prediction model library, establish the board lifespan prediction model through drag-and-drop modeling. In the board mechanism model library, establish the board mechanism model through drag-and-drop modeling. The digital twin lifetime prediction model library includes a model library selection section and a work area section. Models are selected by dragging and dropping them from the model library to the work area section. In the work area section, the model is selected and the model parameter setting interface is entered to set the model parameters. The set model can be copied and pasted for reuse of models with the same parameters. The board mechanism model library includes a model library area and a working area. After dragging the model from the model library area to the working area, the parameters are set, and the components in the working area are connected according to the same connection relationship as the real board to complete the modeling. During the modeling process, the pins of two electronic components are connected by copper plating. Step 2: Read the historical analysis data of the board and calculate the lifespan of the board and vulnerable components in real time through the lifespan prediction model of the board. When there is fault information in the lifespan of the board and vulnerable components, analyze the phenomenon of potential fault units failing through the mechanism model of the board. Otherwise, obtain the analysis data of the board and the expected lifespan of the board. Step 3: In the mechanism model of the board, determine whether the potential fault information will endanger the operation of the entire system. If it will endanger the operation of the entire system, output the expected fault phenomenon of the board with an alarm. Otherwise, output the expected usage plan to the mechanism model of the board to complete the establishment of the digital twin of the circuit board.

2. The method of claim 1, wherein, In step 1, the digital twin lifetime prediction model library includes electronic components, contact devices, and PCB boards. Among them, electronic components include, but are not limited to, semiconductor integrated circuits such as resistors, capacitors, inductors, and light-emitting diodes. Contact devices include, but are not limited to, switches, mechanical relays, and connectors; PCBs include, but are not limited to, FR-4, FPC, aluminum substrate, thermoelectrically separated copper substrate, Rogers high-frequency board, PTFE Teflon high-frequency board and circuit board; The board mechanism model library includes, but is not limited to, electronic devices and connection devices, among which electronic devices include resistors, capacitors, inductors and semiconductor devices; Connection devices include connectors, solder joints, and copper plating on PCBs.

3. The method of claim 1, wherein the method further comprises: In step 1, the digital twin life prediction model library processes temperature, humidity, and pressure to establish independent analysis models for different types of objects and sets parameters individually. The independent analysis models have the function of inserting historical data.

4. The method of claim 1, wherein, In step 1, the board life prediction model sets multiple confidence levels for the board life analysis process and outputs the predicted life under different confidence levels. The expected life curve is compared with the confidence level to obtain the expected life of the board. The confidence level is determined by the quality of the components used in the board and the degree of trust in the board.

5. The method of claim 1, wherein, In step 2, the board's lifespan prediction model predicts the board's lifespan and vulnerable components. The specific lifespan prediction process is as follows: The model checks whether the board has historical data. If there is historical data, the model needs to set the initial lifespan data to the historical data during calculation; otherwise, the model sets the lifespan data to 100 before calculation. After the historical data is analyzed, the board's lifespan prediction model incorporates the data read by the sensors into the algorithm, and calculates the board's wear based on the temperature, humidity, and pressure read by the sensors.

6. The method for establishing a digital twin of a circuit board according to claim 1, characterized in that, In step 2, when there is fault information regarding the lifespan of the board and vulnerable components, the fault phenomenon of the shortest lifespan element is calculated through the board's mechanism model.

7. A system for creating a digital twin of a circuit board, characterized in that, A method for establishing a circuit board digital twin according to any one of claims 1-6 includes: The model building module is used to build a digital twin lifespan prediction model library and a board mechanism model library. Within the digital twin lifespan prediction model library, the lifespan prediction model of the board is built through drag-and-drop modeling. Within the board mechanism model library, the mechanism model of the board is built through drag-and-drop modeling. The lifespan prediction module is used to read historical analysis data of the board and calculate the lifespan of the board and vulnerable components in real time through the lifespan prediction model of the board. When there is fault information in the lifespan of the board and vulnerable components, the mechanism model of the board is used to analyze the phenomenon of potential fault units failing. Otherwise, the analysis data of the board and the expected lifespan of the board are obtained. The fault analysis module is used to determine whether potential fault information in the board's mechanism model will jeopardize the operation of the entire system. If it does, an alarm is outputting the expected fault phenomenon of the board; otherwise, the expected usage plan is output to the board's mechanism model to complete the establishment of the digital twin of the circuit board.

8. A mobile terminal, wherein a computer-readable storage medium stores a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for establishing a circuit board digital twin as described in any one of claims 1-6.

Images