Circuit board service life prediction method based on stress and weak point monitoring

Abstract

The invention relates to the technical field of circuit board fault prediction, in particular to a circuit board service life prediction method based on stress and weak point monitoring, and mainly solves the technical problem that an existing circuit board service life prediction monomer is low in prediction accuracy. According to the circuit board life prediction method based on stress and weak point monitoring, the residual life of a circuit board is calculated through the following steps: S11, carrying out a short-term aging experiment, and collecting early-stage failure rate data; s12, predicting later failure rate data of the device through an autoregressive differential moving average algorithm; s13, establishing a fatigue accumulated loss failure model of the device; s14, calculating the residual life of the circuit board; the welding spot state is judged through the following steps that S21, thermal stress and mechanical stress are analyzed; s22, weak points of connection failure are determined; and S23, establishing a welding spot failure judgment model, and judging the welding spot state of the weak spot. The method can be used in a single body or in batch, and the accuracy of single body prediction can be ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of circuit board failure prediction, and in particular to a circuit board life prediction method based on stress and weak point monitoring. Background Art

[0002] With the rapid expansion of the operation scale of my country's high-speed railways and urban rail transit, the entire rail transit industry has begun to show huge rigid demands for ensuring operational safety, improving service quality and reducing operating costs, and at the same time, has put forward higher and higher requirements for the reliability, availability, maintainability and safety of facilities and equipment. The transition from regular maintenance to condition-based maintenance and predictive maintenance is the development need for rail transit converters to improve the operational efficiency of the entire life cycle and reduce costs. Electronic devices have become the focus of research on prediction because of their large number of devices, complex circuit structure and serious impact after failure.

[0003] The bathtub curve is currently used to predict circuit board failures. The bathtub curve is shaped like a bathtub. Figure 3 As shown in the figure, it has three obvious regions: Region I represents the early failure period, where the failure rate gradually decreases from the highest value; Region II represents the accidental failure period, where the failure rate hardly changes; Region III represents the wear and tear failure period, where the failure rate shows an upward trend. The bathtub curve is a theoretical curve based on experience, which focuses on the overall characteristics of the long-term failure rate of batch products, but has a low accuracy rate in predicting single failures and cannot support individual reliability life research and prediction. Summary of the invention

[0004] In order to overcome the technical defect of low single prediction accuracy in the existing circuit board life prediction, the present invention provides a circuit board life prediction method based on stress and weak point monitoring.

[0005] The circuit board life prediction method based on stress and weak point monitoring provided by the present invention comprises:

[0006] Calculate the remaining life of the circuit board by following the steps below:

[0007] S11. Perform short-term aging experiments on each device and collect early failure rate data;

[0008] S12. Predicting the later failure rate data of the device by using an autoregressive differential moving average algorithm, thereby obtaining the failure rate data of the device throughout its life cycle to form a data set;

[0009] S13. Establishing a device fatigue cumulative loss failure model based on the data set;

[0010] S14. Calculate the remaining life of the circuit board through the device fatigue cumulative loss failure model according to the actual stress conditions during product operation;

[0011] Determine the solder joint status by following the steps below:

[0012] S21. Analyze thermal and mechanical stresses through simulation software modeling;

[0013] S22. Determine the weak point of connection failure;

[0014] S23. Establish a solder joint failure determination model, and determine the solder joint status of the weak point by using the solder joint failure determination model.

[0015] Optionally, in step S12, the autoregressive difference moving average algorithm includes an autoregressive model, a difference process and a moving average model;

[0016] The expression of the autoregressive model is:

[0017] ;

[0018] in, express Time series data at time points, is a constant, Indicates the relationship between current value and past The model parameters for the relationship between observations at time points, express The error term at the time point;

[0019] The difference process includes first-order or second-order difference processing;

[0020] The expression of the moving average model is:

[0021] ;

[0022] in, is a constant, Indicates the current value and the past The model parameters for the relationship between the errors at different time points.

[0023] Optionally, in step S13, the expression of the device fatigue cumulative loss failure model is:

[0024] ;

[0025] in, Indicates the loss degree, Indicated in The actual operating time of the circuit board under each stress condition, Indicates The theoretical life of the circuit board under various stress conditions. , Indicated in The failure rate of the circuit board under each stress condition, , Indicates the total number of device types. Indicates The number of devices of this type, Indicates The failure rate of this type of device.

[0026] Optionally, in step S14, the remaining life of the circuit board is calculated as follows:

[0027] ;

[0028] in, Indicates the remaining life of the circuit board, Indicates the design service life of the circuit board.

[0029] Optionally, in step S23, the solder joint at the weak point is equivalent to a resistor, and capacitors are connected in series to form an equivalent circuit, thereby forming the solder joint failure determination model, and determining the solder joint state by detecting the voltage across the capacitor.

[0030] The technical solution provided by the present invention has the following advantages compared with the prior art:

[0031] The circuit board life prediction method based on stress and weak point monitoring provided by the present invention can calculate the remaining life of the circuit board through the device fatigue cumulative loss failure model, and can determine the solder joint state through the solder joint failure judgment model, so that the device failure and solder joint failure can be predicted at the same time, thereby completing the life prediction of the circuit board. Compared with the existing method of using the bathtub curve for fault prediction, this method can be used for both single units and batches, and can ensure the accuracy of single unit prediction. In addition, this method uses an autoregressive differential moving average algorithm to predict the later failure rate data of the device, which can save time and test costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0034] Figure 1 A flow chart showing a circuit board life prediction method according to an embodiment of the present invention.

[0035] Figure 2 It shows a failure rate prediction verification diagram in an embodiment of the present invention;

[0036] Figure 3 A schematic diagram showing a bathtub curve in the prior art. DETAILED DESCRIPTION

[0037] In order to more clearly understand the above-mentioned objectives, features and advantages of the present invention, the scheme of the present invention will be further described below. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.

[0038] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present invention, rather than all of the embodiments.

[0039] Combine the following Figure 1 The specific embodiments of the present invention are described in detail.

[0040] This embodiment provides a circuit board life prediction method based on stress and weak point monitoring, which includes two parts: one part is to calculate the remaining life of the circuit board, and the other part is to determine the state of the solder joint.

[0041] The remaining life of the circuit board is calculated through steps S11 to S15.

[0042] S11. Perform short-term aging experiments on each device and collect preliminary failure rate data.

[0043] It is easy to understand that long-term testing will cost a lot of time and money, so this method uses short-term aging tests to obtain early failure rate data, and then the later failure rate data is obtained through prediction.

[0044] S12. Predict the later failure rate data of the device through the autoregressive differential moving average algorithm, so as to obtain the failure rate data of the device throughout its life cycle and form a data set.

[0045] Specifically, the autoregressive difference moving average algorithm includes an autoregressive model, a difference process and a moving average model;

[0046] The expression of the autoregressive model is:

[0047] ;

[0048] in, express Time series data at time points, is a constant, Indicates the relationship between current value and past The model parameters for the relationship between observations at time points, express The error term at the time point;

[0049] The difference process includes first-order or second-order difference processing;

[0050] The expression of the moving average model is:

[0051] ;

[0052] in, is a constant, Indicates the relationship between current value and past The model parameters for the relationship between the errors at different time points.

[0053] It should be noted that the autoregressive model is used to process the autoregressive part of the time series, taking into account the impact of observations from several past periods on the current value; the differential part is used to make the non-stationary time series stable, and eliminate the trend and seasonal factors in the time series through first-order or second-order differential processing; the moving average model is used to process the moving average part of the time series, taking into account the impact of past prediction errors on the current value. Through these three parts, the autoregressive differential moving average algorithm can capture the trend changes of the data and handle those temporary, sudden changes or data with large noise. Input the previous short-term data and use the autoregressive differential moving average algorithm to predict the failure rate of the device throughout the life cycle.

[0054] Figure 2 In the figure, the black part is the aging test data, and the purple part is the long-term aging test data predicted by the autoregressive difference moving average algorithm. It can be seen from the figure that the simulation effect is good. Through this algorithm, the short-term aging test data of the device can be simulated into the full life cycle data.

[0055] S13. Establish a device fatigue cumulative loss failure model based on the data set.

[0056] Specifically, the expression of the device fatigue cumulative loss failure model is:

[0057] ;

[0058] in, Indicates the loss degree, Indicated in The actual operating time of the circuit board under each stress condition, Indicates The theoretical life of the circuit board under various stress conditions. , Indicated in The failure rate of the circuit board under each stress condition, , Indicates the total number of device types. Indicates The number of devices of this type, Indicates The failure rate of this type of device.

[0059] It should be noted that if there are S stress conditions on the circuit board in theory, the full life model of the circuit board is:

[0060] .

[0061] S14. According to the actual stress conditions during product operation, the remaining life of the circuit board is calculated using the device fatigue cumulative loss failure model.

[0062] Specifically, the calculation formula for the remaining life of the circuit board is:

[0063] ;

[0064] in, Indicates the remaining life of the circuit board, Indicates the design service life of the circuit board.

[0065] The state of the soldering point is determined through steps S21 to S23.

[0066] It should be noted that circuit board failure is divided into two categories: device failure and solder joint failure. Device failure is mainly related to the failure rate characteristics related to the device composition characteristics and the stress during use, while soldering failure is mainly related to factors such as structural tension and heat generation during the use of solder joints.

[0067] S21. Analyze thermal and mechanical stresses through modeling with simulation software.

[0068] S22. Determine weak points of connection failure. For example, in this embodiment, it is found that the solder joints located at the four corners of the BGA packaged FPGA are the first to fail under environmental stress, so these four corners are used as weak points for detection.

[0069] It should be noted that it is generally impossible to monitor all components in the entire system at the same time when performing fault diagnosis. Through simulation, engineering experiments or high-acceleration experiments, the components of a system that are most prone to failure in the working environment can be determined. The working status of these sensitive components can reflect the degree of accumulation of various environmental stresses in the environment on the system. By using these few sensitive components to establish fault diagnosis points, the working status of the system can be effectively judged and the working life of the system can be predicted.

[0070] S23. Establish a solder joint failure determination model, and determine the solder joint status of the weak point through the solder joint failure determination model.

[0071] Specifically, the solder joint at the weak point is equivalent to a resistor, and a capacitor is connected in series to form an equivalent circuit, thereby forming a solder joint failure determination model, and the solder joint state is determined by detecting the voltage across the capacitor.

[0072] It is easy to understand that the resistance of an object is proportional to the length of the object and inversely proportional to the cross-sectional area of ​​the object. The resistance of an object per unit length and per unit cross-sectional area is called the resistivity of the object. The formula is as follows:

[0073] ;

[0074] in, represents the resistance of the object, represents the resistivity of the object, Indicates the length of an object. Represents the cross-sectional area of ​​an object.

[0075] It is easy to understand that in the equivalent circuit, the voltage has the following relationship:

[0076] ;

[0077] in, Indicates the power supply voltage, Represents the equivalent resistance voltage of the solder joint, Represents the voltage of capacitors connected in series.

[0078] The expression for the capacitor voltage is:

[0079] ;

[0080] in, is the time constant of the equivalent circuit, The larger the value, the longer the charging time required, and the slower the transient process progresses. The core of the BIT circuit in this embodiment works in the FPGA to be tested, and determines whether there is a failure and the type of failure by reading the capacitor voltage signal. Since the FPGA input and output ports are digital signals rather than analog signals, there is a certain threshold voltage, and the voltage signal read can only be displayed as a high level "1" or a low level "0". In a cycle test, the BIT first writes logic 1 and charges the capacitor, and then reads the voltage of the charged capacitor. If the solder joint network is not damaged, writing logic 1 will fully charge the capacitor and a logic 1 will be read. When the solder joint network is completely damaged, the time constant becomes larger, so that the capacitor is not fully charged, and the amplitude of the corresponding capacitor voltage signal will decrease. When the capacitor voltage is lower than the threshold voltage of the FPGA input and output pins, a logic 0 will be read, indicating that a solder joint network has been damaged at the solder joint, and a high resistance failure that is severe enough and lasts long enough has occurred, which leads to the failure of the write operation.

[0081] The overall idea of ​​the circuit board life prediction method based on stress and weak point monitoring in this embodiment is as follows:

[0082] First, conduct short-term aging tests on typical components of circuit boards, and use the autoregressive differential moving average model and short-term aging test data to simulate the loading of the components' respective full life cycle data; use the fatigue cumulative loss model of electronic product components and the full life cycle data set of the components to establish a circuit board component fatigue cumulative loss failure model; and calculate the remaining life of the circuit board through the circuit board component fatigue cumulative loss failure model based on the stress monitoring data set collected during product operation.

[0083] Second, conduct thermal stress and mechanical stress simulation modeling on the circuit board, determine the weak points of connection failure on the circuit board through simulation, monitor the weakest points, connect a charging capacitor equivalent circuit in series, and build a circuit board solder joint failure determination model; determine the solder joint status through the voltage signal during operation. If the solder joint status is poor, the circuit board status needs to be replaced or repaired.

[0084] Circuit board status monitoring and life prediction are achieved through one and two.

[0085] The above is only a specific implementation of the present invention, which enables those skilled in the art to understand or implement the present invention. Although detailed descriptions are given with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments, and they should all be covered by the protection scope of the claims.

Claims

1. A circuit board life prediction method based on stress and weak point monitoring, characterized in that: Calculate the remaining life of the circuit board by following the steps below: S11. Perform short-term aging experiments on each device and collect early failure rate data; S12. Predicting the later failure rate data of the device by using an autoregressive differential moving average algorithm, thereby obtaining the failure rate data of the device throughout its life cycle to form a data set; S13. Establishing a device fatigue cumulative loss failure model based on the data set; S14. Calculate the remaining life of the circuit board through the device fatigue cumulative loss failure model according to the actual stress conditions during product operation; Determine the solder joint status by following the steps below: S21. Analyze thermal and mechanical stresses through simulation software modeling; S22. Determine the weak point of connection failure; S23. Establish a solder joint failure determination model, and determine the solder joint status of the weak point by using the solder joint failure determination model.

2. The circuit board life prediction method based on stress and weak point monitoring according to claim 1 is characterized in that: In step S12, the autoregressive difference moving average algorithm includes an autoregressive model, a difference process and a moving average model; The expression of the autoregressive model is: ； in, express Time series data at time points, is a constant, Indicates the current value and the past The model parameters for the relationship between observations at time points, express The error term at the time point; The difference process includes first-order or second-order difference processing; The expression of the moving average model is: ； in, is a constant, Indicates the current value and the past The model parameters for the relationship between the errors at different time points.

3. The circuit board life prediction method based on stress and weak point monitoring according to claim 1 is characterized in that: In step S13, the expression of the device fatigue cumulative loss failure model is: ； in, Indicates the loss degree, Indicated in The actual operating time of the circuit board under each stress condition, Indicates The theoretical life of the circuit board under various stress conditions. , Indicated in The failure rate of the circuit board under each stress condition, , Indicates the total number of device types. Indicates The number of devices of this type, Indicates The failure rate of this type of device.

4. The circuit board life prediction method based on stress and weak point monitoring according to claim 3 is characterized in that: In step S14, the remaining life of the circuit board is calculated as follows: ； in, Indicates the remaining life of the circuit board, Indicates the design service life of the circuit board.

5. The circuit board life prediction method based on stress and weak point monitoring according to claim 1 is characterized in that: In step S23, the solder joint at the weak point is equivalent to a resistor, and a capacitor is connected in series to form an equivalent circuit, thereby forming the solder joint failure determination model, and determining the solder joint state by detecting the voltage across the capacitor.

Images