Method for analyzing fusing capability of high-current gold bonding wire

Abstract

The invention discloses a high-current gold bonding wire fusing capability analysis method and system, and the method comprises the steps: carrying out the preprocessing of collected multi-dimensional data and waveform feature data; constructing a fusing prediction model, inputting the preprocessed multi-dimensional data and waveform feature data into the fusing prediction model, and predicting the fusing probability of the gold bonding wire; based on the fusing probability of the gold bonding wire, fusing factors are analyzed, and fusing protection is carried out on the gold bonding wire through self-adaptive adjustment and an alarm mechanism; after fusing prediction and protection measures are completed, a fusing capacity analysis report is generated, and the fusing prediction model is optimized. According to the method, the accuracy and the stability of the prediction model are remarkably improved by comprehensively collecting the multi-dimensional data and the waveform feature data and carrying out preprocessing, feature extraction and model construction; and by analyzing fusing factors and implementing a self-adaptive adjustment and alarm mechanism, the gold bonding wire can be ensured to be in a safe working range under various working conditions.

Description

Technical Field

[0001] The invention relates to the fields of electronic engineering and material science, and in particular to a method and system for analyzing the fusing capacity of a large current bonding gold wire. Background Art

[0002] With the rapid development of modern electronic technology, high-density, high-performance electronic components are becoming more and more common in various applications. As one of the key components in semiconductor packaging, it is crucial to ensure the fusing capacity of gold bonding wire. Gold bonding wire is mainly used for electrical connection between chips and external pins. Its reliability and durability directly affect the performance and life of the entire electronic device. In recent years, with the increasing trend of miniaturization and integration of electronic equipment, the requirements for its fusing capacity have been increasing. However, there is relatively little research on the analysis of the fusing capacity of gold bonding wire, especially the fusing mechanism and its prediction method are not yet mature.

[0003] Traditional methods for analyzing the fusing capacity of gold bonding wires usually rely on static experimental data and empirical formulas. Although this method is simple and easy to use, it has obvious limitations in practical applications. First, static experimental data are difficult to reflect the true behavior of gold bonding wires under dynamic working conditions. In particular, when factors such as current intensity, temperature fluctuations, and stress levels change, the actual fusing probability of gold bonding wires also changes. Second, existing empirical formulas often ignore the complex interactions between multidimensional features, resulting in inaccurate prediction results. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method and system for analyzing the fusing capacity of a large current bonding gold wire to solve the problem of inaccurate prediction in the prior art.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, an embodiment of the present invention provides a method for analyzing the fusing capacity of a large current bonding gold wire, which includes: collecting multi-dimensional data and waveform characteristic data through a data acquisition device; preprocessing the collected multi-dimensional data and waveform characteristic data; constructing a fusing prediction model, inputting the preprocessed multi-dimensional data and waveform characteristic data into the fusing prediction model, and predicting the fusing probability of the bonding gold wire; based on the fusing probability of the bonding gold wire, analyzing the fusing factors, and performing fusing protection on the bonding gold wire through adaptive adjustment and alarm mechanisms; after completing the fusing prediction and protection measures, generating a fusing capacity analysis report, and optimizing the fusing prediction model.

[0008] As a preferred solution for analyzing the fusing capacity of the high current bonding wire of the present invention, wherein:

[0009] The specific steps of collecting multi-dimensional data and waveform characteristic data are as follows:

[0010] Collect waveform characteristic data through oscilloscope, current probe and voltage probe;

[0011] Multi-dimensional data is collected through current sensors, voltage sensors, temperature sensors and stress sensors.

[0012] As a preferred solution for analyzing the fusing capacity of the high current bonding wire of the present invention, the collected multi-dimensional data and waveform characteristic data are pre-processed, and the specific steps are as follows:

[0013] Use low-pass filters to remove high-frequency noise, retain low-frequency signals, and reduce interference factors in the data;

[0014] Eliminate extreme values ​​in data through anomaly detection and removal functions;

[0015] Through normalization processing, the multi-dimensional data and waveform characteristics collected by the data acquisition device are unified, and the expression is:

[0016]

[0017] Where x is the original data value, x′ is the normalized data value, max(x) is the maximum value in the original data, min(x) is the minimum value in the original data, and the original data value is the value obtained from the data acquisition device without any processing before data preprocessing;

[0018] The stability of the fuse prediction model is improved through standardization. The standardization expression is:

[0019]

[0020] Among them, x is the original data value, z is the standardized data value, μ is the original data mean, and σ is the standard deviation of the original data;

[0021] By feature extraction, the RMS value and peak value of the current, the amplitude and phase of the harmonics of the current waveform, the maximum value, minimum value and average value of the temperature feature, and the maximum value, minimum value and average value of the stress feature are extracted;

[0022] The expression of the RMS value of the current waveform is:

[0023]

[0024] Among them, I RMS is the RMS value of the current, I i is the current value of the i-th sampling point, and N is the total number of sampling points;

[0025] The amplitude and phase of the harmonics of the current waveform are extracted using Fourier transform, and the expression is:

[0026]

[0027] Among them, H k is the amplitude of the kth harmonic, I n is the current value at the nth sampling point, and j is an imaginary unit.

[0028] As a preferred solution for analyzing the fusing ability of the high current bonding gold wire of the present invention, the fusing prediction model is constructed, and the pre-processed multi-dimensional data and waveform characteristic data are input into the fusing prediction model to predict the fusing probability of the bonding gold wire. The specific steps are as follows:

[0029] The logistic regression model is selected as the basic model, polynomial features and cross terms are introduced, and the nonlinear relationship between features is captured using polynomial features and cross terms. The collected multi-dimensional data and waveform feature data are divided into training set, test set and validation set.

[0030] The training set data is used to train the logistic regression model, and the validation set data is used to validate the model, adjust the parameters, and identify the optimal model parameters to construct a model for predicting the melting probability of the bonding wire. The expression is:

[0031]

[0032] Among them, P(y=1|y) is the probability of the bonding wire melting, y is the input feature vector, α i is the regression coefficient of the ith feature, g i (y i ) is y i The transformation function of y i is the characteristic measurement value, c is the intercept term;

[0033] The model performance is evaluated using the test set, and the fuse prediction expression is used to calculate the fuse probability of each test sample.

[0034] As a preferred solution for analyzing the fusing capacity of the high current bonding wire of the present invention, the fusing factors are analyzed based on the fusing probability, and the specific steps are as follows:

[0035] According to the probability of the bonding wire fusing, the correlation coefficient between each feature and the probability of the bonding wire fusing is calculated, and the recursive feature elimination method is used to eliminate unimportant features, and the key features that affect the probability of fusing are found. The key features are controlled within the range that will not have a significant impact on the probability of fusing through the control variable method. The recursive feature elimination method is used again to eliminate unimportant features, and the key features that have the greatest impact on the probability of the bonding wire fusing are gradually found. The key features are sorted from high to low in the order in which they are found, and a correlation feature queue is generated.

[0036] As a preferred solution for analyzing the fusing capacity of the high current gold bonding wire of the present invention, the fusing protection of the gold bonding wire is performed through the adaptive adjustment and alarm mechanism, and the specific steps are as follows:

[0037] A dynamic adjustment strategy is used to monitor the state of the bonding wire in real time. The dynamic threshold is set as T(t) according to the working state. The expression is:

[0038]

[0039] Among them, α is the weight coefficient of the first set of features, β is the weight coefficient of the second set of features, which is used to balance the importance of different features, and w i and v i are the weights of the i-th feature, x i (t) and y i (t) represents the current characteristic and environmental characteristic values ​​at time t, n and m are the number of current characteristics and environmental characteristics, respectively;

[0040] During the monitoring process, the bonding wire status index is calculated according to the status index function and compared with the dynamically adjusted threshold. The status index function expression is:

[0041]

[0042] Among them, S(t) is the state index, γ i is the i-th characteristic coefficient, z i (t) is the i-th eigenvalue at t, t is any time, K i is the saturation constant of the i-th feature, p is the number of features, and δ is the bias term;

[0043] The state index of the bonding wire is calculated according to the state index function, and the dynamic threshold T(t) is compared with the state index S(t), and the adaptive adjustment mechanism is set.

[0044] When S(t)>T(t), the bonding wire is in a dangerous state, and the features in the correlation feature queue are gradually adjusted starting from the feature that has the greatest impact on the probability of the bonding wire fusing;

[0045] When S(t)≤T(t), the working state of the bonding wire is still within the safe range, and the working parameters are optimized to improve efficiency and reduce costs;

[0046] When the status indicator S(t) exceeds the dynamic threshold T(t), the alarm mechanism is triggered and a warning signal is sent to notify the operator to take necessary actions.

[0047] As a preferred solution for analyzing the fusing capacity of the high current bonding wire of the present invention, after completing the fusing prediction and protection measures, a fusing capacity analysis report is generated, and the fusing prediction model is optimized. The specific steps are as follows:

[0048] Generate a fusing capacity analysis report based on the predicted fusing probability, key influencing factors and protection measures;

[0049] Based on the information in the fuse capacity analysis report, the model parameters are adjusted and the feature selection is optimized to iteratively optimize the fuse prediction model.

[0050] In the second aspect, the present invention provides a method and system for analyzing the fusing capacity of a large current bonding gold wire, including: a data acquisition module, which is used to collect multi-dimensional data and waveform characteristic data through a data acquisition device; a data preprocessing module, which is used to preprocess the collected multi-dimensional data and waveform characteristic data; a fusing prediction model construction module, which is used to construct a fusing prediction model, input the preprocessed multi-dimensional data and waveform characteristic data into the fusing prediction model, and predict the fusing probability of the bonding gold wire; a fusing protection module, which is used to analyze the fusing factors based on the fusing probability of the bonding gold wire, and perform fusing protection on the bonding gold wire through adaptive adjustment and alarm mechanism; an analysis report generation module, which is used to generate a fusing capacity analysis report and optimize the fusing prediction model after completing the fusing prediction and protection measures.

[0051] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for analyzing the fusing capacity of a high current bonding gold wire as described in the first aspect of the present invention is implemented.

[0052] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for analyzing the fusing capacity of a high current bonding gold wire as described in the first aspect of the present invention is implemented.

[0053] The beneficial effects of the present invention are as follows: by comprehensively collecting multi-dimensional data and waveform feature data, and performing preprocessing, feature extraction and model construction, the accuracy and stability of the prediction model are significantly improved; by analyzing the fusing factors and implementing adaptive adjustment and alarm mechanisms, it is ensured that the bonding wire is within the safe working range under various working conditions. The present invention belongs to the field of electronic engineering and material science, and is particularly suitable for electronic devices and circuits that require high-reliability connections, and aims to improve the analysis accuracy and protection effect of the fusing capacity of the bonding wire under high current conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on the accompanying drawings without paying any creative work.

[0055] Figure 1 This is a flow chart of the method for analyzing the fusing capacity of a high current bonding gold wire in Example 1.

[0056] Figure 2 This is a flow chart of the system for analyzing the fusing capacity of a high current bonding gold wire in Example 1. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0058] 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, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0059] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0060] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a method for analyzing the fusing capacity of a large current bonding wire, comprising the following steps:

[0061] S1: Collect multi-dimensional data and waveform characteristic data through data acquisition equipment;

[0062] Furthermore, waveform characteristic data is collected by using an oscilloscope, current probe, and voltage probe;

[0063] Multi-dimensional data is collected through current sensors, voltage sensors, temperature sensors and stress sensors. It should be noted that the current and voltage waveform changes of the bonding wire under dynamic working conditions can be obtained in real time to more accurately reflect its actual working state.

[0064] S2: preprocessing the collected multi-dimensional data and waveform feature data;

[0065] Furthermore, high-frequency noise is removed through a low-pass filter, low-frequency signals are retained, and interference factors in the data are reduced;

[0066] For example, electromagnetic interference (EMI), power supply noise, signal fluctuations caused by mechanical vibration, and other high-frequency transient signals.

[0067] Eliminate extreme values ​​in data through anomaly detection and removal functions;

[0068] For example, abnormally high current peaks, sudden temperature changes, or sudden stress increases due to measurement errors, equipment failures, or other abnormal conditions.

[0069] Through normalization processing, the multi-dimensional data and waveform characteristics collected by the data acquisition device are unified, and the expression is:

[0070]

[0071] Where x is the original data value, x′ is the normalized data value, max(x) is the maximum value in the original data, min(x) is the minimum value in the original data, and the original data value is the value obtained from the data acquisition device without any processing before data preprocessing;

[0072] It should be noted that normalization can eliminate the dimensional differences between different features and make different features comparable on the same scale, which is crucial for subsequent model training; normalization can improve the convergence speed and stability of the model, especially when using the gradient descent method for optimization, the normalized data can converge to the optimal solution faster.

[0073] The stability of the fuse prediction model is improved through standardization. The standardization expression is:

[0074]

[0075] Among them, x is the original data value, z is the standardized data value, μ is the original data mean, and σ is the standard deviation of the original data;

[0076] It should be noted that raw data refers to data obtained directly from the data acquisition device without any processing; after standardized processing, the data will be compared on the same scale to facilitate subsequent data analysis and model training; if there are extreme values ​​in the data set, standardization may cause some data points to deviate far from the center. In this case, it may be necessary to first process the outliers and then standardize.

[0077] By feature extraction, the RMS value and peak value of the current, the amplitude and phase of the harmonics of the current waveform, the maximum value, minimum value and average value of the temperature feature, and the maximum value, minimum value and average value of the stress feature are extracted;

[0078] The expression of the RMS value of the current waveform is:

[0079]

[0080] Among them, I RMS is the RMS value of the current, I i is the current value of the i-th sampling point, and N is the total number of sampling points;

[0081] It should be noted that the RMS value is an important statistic that can be used to measure the energy level of the current waveform. If the current waveform has obvious periodicity, sampling should be performed within a complete cycle to obtain an accurate RMS value. The total number of sampling points N should be large enough to ensure the accuracy of the calculation results.

[0082] For example, suppose we have a current waveform, the total number of sampling points N = 10, and the sampling point currents are I1, I2, ..., I10, then the RMS value I RMS The calculation can be performed according to the following steps: first calculate the sum of the squares of the currents, then calculate the average value, and finally take the square root of the average value to obtain the RMS value of the current.

[0083] The amplitude and phase of the harmonics of the current waveform are extracted using Fourier transform, and the expression is:

[0084]

[0085] Among them, H k is the amplitude of the kth harmonic, I n is the current value at the nth sampling point, and j is an imaginary unit.

[0086] It should be noted that Fourier transform is a mathematical tool used to convert a signal in the time domain into a representation in the frequency domain.

[0087] S3: construct a fusing prediction model, input the pre-processed multi-dimensional data and waveform feature data into the fusing prediction model, and predict the fusing probability of the bonding wire;

[0088] Furthermore, the logistic regression model is selected as the basic model, polynomial features and cross terms are introduced, and the nonlinear relationship between features is captured using polynomial features and cross terms. The collected multi-dimensional data and waveform feature data are divided into training set, test set and validation set;

[0089] It should be noted that the polynomial feature generator is used to automatically create polynomial features. Note that randomness should be ensured when dividing the data set to prevent distribution bias between data sets.

[0090] The training set data is used to train the logistic regression model, and the validation set data is used to validate the model, adjust the parameters, and identify the optimal model parameters to construct a model for predicting the melting probability of the bonding wire. The expression is:

[0091]

[0092] Among them, P(y=1|y) is the probability of the bonding wire melting, y is the input feature vector, α i is the regression coefficient of the ith feature, g i (y i ) is y i The transformation function of y i is the characteristic measurement value, c is the intercept term;

[0093] It should be noted that the eigenvector y contains all the eigenvalues ​​used for prediction, and the regression coefficient α i It reflects the influence of the features on the probability of melting of the bonding wire. The logistic regression model can not only effectively handle the prediction of melting of the bonding wire under dynamic working conditions, but also capture the nonlinear relationship between the features by introducing polynomial features and cross terms, thereby improving the accuracy and reliability of the prediction.

[0094] Use the test set to evaluate the model performance and use the circuit breaker prediction expression to calculate the circuit breaker probability of each test sample.

[0095] For example, assuming the feature vector y, substitute it into the expression for predicting the melting probability of the bonding wire, calculate the melting probability of each test sample, and evaluate the model performance based on the calculation results.

[0096] According to the probability of the bonding wire fusing, the correlation coefficient between each feature and the probability of the bonding wire fusing is calculated, and the recursive feature elimination method is used to eliminate unimportant features, and the key features that affect the probability of fusing are found. The key features are controlled within the range that will not have a significant impact on the probability of fusing through the control variable method. The recursive feature elimination method is used again to eliminate unimportant features, and the key features that have the greatest impact on the probability of the bonding wire fusing are gradually found. The key features are sorted from high to low in the order in which they are found, and a correlation feature queue is generated.

[0097] It should be noted that by calculating the correlation coefficient between each feature and the probability of melting of the bonding wire, the strength of the linear relationship between the feature and the probability of melting can be quantified. The larger the absolute value of the correlation coefficient, the more significant the impact of the feature on the probability of melting. Recursive feature elimination is a feature selection method that finds the optimal feature subset by recursively removing the least relevant features.

[0098] S4: Based on the melting probability of the bonding wire, the melting factors are analyzed, and the bonding wire is protected from melting through adaptive adjustment and alarm mechanism;

[0099] Furthermore, a dynamic adjustment strategy is used to monitor the state of the bonding wire in real time, and the dynamic threshold is set as T(t) according to the working state. The expression is:

[0100]

[0101] Among them, α is the weight coefficient of the first set of features, β is the weight coefficient of the second set of features, which is used to balance the importance of different features, and w i and v i are the weights of the i-th feature, x i (t) and y i (t) represents the current characteristic and environmental characteristic values ​​at time t, n and m are the number of current characteristics and environmental characteristics, respectively;

[0102] It should be noted that the setting of the dynamic threshold T(t) comprehensively considers the importance of current characteristics and environmental characteristics. α is the weighting coefficient of the first group of characteristics (usually current characteristics), which is used to balance the importance of current characteristics. β is the weighting coefficient of the second group of characteristics (usually environmental characteristics), which is used to balance the importance of environmental characteristics. The role of the feature weight is to reflect the degree of influence of the feature on the state of the bonding wire. The dynamic threshold can be used to monitor the state of the bonding wire in real time, and adjust the working parameters according to actual conditions to ensure that the bonding wire is always within a safe working range.

[0103] During the monitoring process, the bonding wire status index is calculated according to the status index function and compared with the dynamically adjusted threshold. The status index function expression is:

[0104]

[0105] Among them, S(t) is the state index, γ i is the i-th characteristic coefficient, z i (t) is the i-th eigenvalue at t, t is any time, K i is the saturation constant of the i-th feature, p is the total number of features, and δ is the bias term;

[0106] It should be noted that the state index S(t) is used to evaluate the state of the bonding wire in real time, reflecting whether the working state of the bonding wire is close to the fusing state at a specific moment. The characteristic coefficient γ i It reflects the influence of this feature on the state of the bonding wire. The characteristic value z i (t) can be current characteristics, temperature characteristics, or stress characteristics.

[0107] The state index of the bonding wire is calculated according to the state index function, and the dynamic threshold T(t) is compared with the state index S(t), and the adaptive adjustment mechanism is set.

[0108] When S(t)>T(t), the bonding wire is in a dangerous state, and the features in the correlation feature queue are gradually adjusted starting from the feature that has the greatest impact on the probability of the bonding wire fusing;

[0109] When S(t)≤T(t), the working state of the bonding wire is still within the safe range, optimizing the working parameters, improving efficiency and reducing costs; it should be noted that the adaptive adjustment mechanism ensures that it is always within the safe working range, thereby improving the reliability and service life of the bonding wire.

[0110] When the status indicator S(t) exceeds the dynamic threshold T(t), the alarm mechanism is triggered and a warning signal is sent to notify the operator to take necessary actions.

[0111] S5: After completing the fusing prediction and protection measures, generate a fusing capacity analysis report and optimize the fusing prediction model;

[0112] Furthermore, a comprehensive forecast of the probability of breaking, key influencing factors and protection measures is used to generate a breaking capacity analysis report;

[0113] Based on the information in the fuse capacity analysis report, the model parameters are adjusted and the feature selection is optimized to iteratively optimize the fuse prediction model.

[0114] It should be noted that the generated fusing capacity analysis report not only provides detailed prediction results and protection measures recommendations, but also iteratively optimizes the fusing prediction model by adjusting model parameters and optimizing feature selection, thereby improving the accuracy and reliability of bonding wire fusing prediction.

[0115] This embodiment also provides a method and system for analyzing the fusing capacity of a high-current bonding gold wire, including: a data acquisition module, for collecting multi-dimensional data and waveform characteristic data through a data acquisition device;

[0116] A data preprocessing module is used to preprocess the collected multi-dimensional data and waveform feature data;

[0117] A fusing prediction model building module is used to build a fusing prediction model, input the pre-processed multi-dimensional data and waveform characteristic data into the fusing prediction model, and predict the fusing probability of the bonding wire;

[0118] The fuse protection module is used to analyze the fuse factors based on the fuse probability of the bonding wire and to perform fuse protection on the bonding wire through adaptive adjustment and alarm mechanism;

[0119] The analysis report generation module is used to generate a fusing capacity analysis report and optimize the fusing prediction model after completing the fusing prediction and protection measures.

[0120] This embodiment also provides a computer device, which is suitable for the method of analyzing the fusing capacity of a large current bonding gold wire, and includes: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the method of analyzing the fusing capacity of a large current bonding gold wire proposed in the above embodiment.

[0121] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0122] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for analyzing the fusing capacity of a large current bonding gold wire proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, disk or optical disk.

[0123] In summary, the present invention includes: using a data acquisition device to acquire multi-dimensional data and waveform characteristic data; preprocessing the acquired multi-dimensional data and waveform characteristic data; constructing a fusing prediction model, inputting the preprocessed multi-dimensional data and waveform characteristic data into the fusing prediction model, and predicting the fusing probability of the bonding gold wire; analyzing the fusing factors based on the fusing probability of the bonding gold wire, and performing fusing protection on the bonding gold wire through adaptive adjustment and alarm mechanisms; after completing the fusing prediction and protection measures, generating a fusing capacity analysis report, and optimizing the fusing prediction model.

[0124] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for analyzing the fusing capacity of a high current gold bonding wire, characterized in that: include, Collect multi-dimensional data and waveform characteristic data through data acquisition equipment; Preprocess the collected multi-dimensional data and waveform characteristic data; Construct a fusing prediction model, input the pre-processed multi-dimensional data and waveform characteristic data into the fusing prediction model, and predict the fusing probability of the bonding wire; Based on the melting probability of the bonding wire, the melting factors are analyzed, and the bonding wire is protected from melting through adaptive adjustment and alarm mechanism; After completing the circuit breaker prediction and protection measures, generate a circuit breaker capacity analysis report and optimize the circuit breaker prediction model.

2. The method for analyzing the fusing capacity of a high current bonding gold wire according to claim 1, characterized in that: The specific steps of collecting multi-dimensional data and waveform characteristic data are as follows: Collect waveform characteristic data through oscilloscope, current probe and voltage probe; Multi-dimensional data is collected through current sensors, voltage sensors, temperature sensors and stress sensors.

3. The method for analyzing the fusing capacity of a high current bonding gold wire according to claim 2, characterized in that: The collected multi-dimensional data and waveform feature data are preprocessed, and the specific steps are as follows: Use low-pass filters to remove high-frequency noise, retain low-frequency signals, and reduce interference factors in the data; Eliminate extreme values ​​in data through anomaly detection and removal functions; Through normalization processing, the multi-dimensional data and waveform characteristics collected by the data acquisition equipment are unified; Through standardization, the stability of the fuse prediction model is improved. The standardization expression is: By feature extraction, the RMS value and peak value of the current, the amplitude and phase of the harmonics of the current waveform, the maximum value, minimum value and average value of the temperature feature, and the maximum value, minimum value and average value of the stress feature are extracted; The expression of the RMS value of the current waveform is: Among them, I RMS is the RMS value of the current, I i is the current value of the i-th sampling point, and N is the total number of sampling points; The amplitude and phase of the harmonics of the current waveform are extracted using Fourier transform, and the expression is: Among them, H k is the amplitude of the kth harmonic, I n is the current value at the nth sampling point, and j is an imaginary unit.

4. The method for analyzing the fusing capacity of a high current bonding gold wire according to claim 3, characterized in that: The fusing prediction model is constructed, and the pre-processed multi-dimensional data and waveform characteristic data are input into the fusing prediction model to predict the fusing probability of the bonding wire. The specific steps are as follows: The logistic regression model is selected as the basic model, polynomial features and cross terms are introduced, and the nonlinear relationship between features is captured using polynomial features and cross terms. The collected multi-dimensional data and waveform feature data are divided into training set, test set and validation set. The training set data is used to train the logistic regression model, and the validation set data is used to validate the model, adjust the parameters, and identify the optimal model parameters to construct a model for predicting the melting probability of the bonding wire. The expression is: Among them, P(y=1|y) is the probability of the bonding wire melting, y is the input feature vector, α i is the regression coefficient of the ith feature, g i (y i ) is y i The transformation function of y i is the characteristic measurement value, c is the intercept term; The model performance is evaluated using the test set, and the fuse prediction expression is used to calculate the fuse probability of each test sample.

5. The method for analyzing the fusing capacity of a high current bonding gold wire according to claim 4, characterized in that: The specific steps of analyzing the fusing factors based on the fusing probability of the bonding wire are as follows: According to the probability of the bonding wire fusing, the correlation coefficient between each feature and the probability of the bonding wire fusing is calculated, and the recursive feature elimination method is used to eliminate unimportant features, and the key features that affect the probability of fusing are found. The key features are controlled within the range that will not have a significant impact on the probability of fusing through the control variable method. The recursive feature elimination method is used again to eliminate unimportant features, and the key features that have the greatest impact on the probability of the bonding wire fusing are gradually found. The key features are sorted from high to low in the order in which they are found, and a correlation feature queue is generated.

6. The method for analyzing the fusing capacity of a high current bonding gold wire according to claim 5, characterized in that: The specific steps of fusing the bonding wire through the adaptive adjustment and alarm mechanism are as follows: A dynamic adjustment strategy is used to monitor the state of the bonding wire in real time. The dynamic threshold is set as T(t) according to the working state. The expression is: Among them, α is the weight coefficient of the first set of features, β is the weight coefficient of the second set of features, which is used to balance the importance of different features, and w i and v i are the weights of the i-th feature, x i (t) and y i (t) represents the current characteristic and environmental characteristic values ​​at time t, n and m are the number of current characteristics and environmental characteristics, respectively; During the monitoring process, the bonding wire status index is calculated according to the status index function and compared with the dynamically adjusted threshold. The status index function expression is: Among them, S(t) is the state index, γ i is the i-th characteristic coefficient, z i (t) is the i-th eigenvalue at t, t is any time, K i is the saturation constant of the i-th feature, p is the number of features, and δ is the bias term; The state index of the bonding wire is calculated according to the state index function, and the dynamic threshold T(t) is compared with the state index S(t), and the adaptive adjustment mechanism is set. When S(t)>T(t), the bonding wire is in a dangerous state, and the features in the correlation feature queue are gradually adjusted starting from the feature that has the greatest impact on the probability of the bonding wire fusing; When S(t)≤T(t), the working state of the bonding wire is still within the safe range, and the working parameters are optimized to improve efficiency and reduce costs; When the status indicator S(t) exceeds the dynamic threshold T(t), the alarm mechanism is triggered and a warning signal is sent to notify the operator to take necessary actions.

7. The method for analyzing the fusing capacity of a high current bonding gold wire according to claim 6, characterized in that: After completing the fuse prediction and protection measures, a fuse capacity analysis report is generated and the fuse prediction model is optimized. The specific steps are as follows: Generate a fusing capacity analysis report based on the predicted fusing probability, key influencing factors and protection measures; Based on the information in the circuit breaker capacity analysis report, the model parameters are adjusted and the feature selection is optimized to iteratively optimize the circuit breaker prediction model.

8. A system for analyzing the fusing capacity of a high current gold bonding wire, based on the analysis of the fusing capacity of a high current gold bonding wire according to any one of claims 1 to 7, characterized in that: include, A data acquisition module, used for collecting multi-dimensional data and waveform characteristic data through a data acquisition device; A data preprocessing module is used to preprocess the collected multi-dimensional data and waveform feature data; A fusing prediction model building module is used to build a fusing prediction model, input the pre-processed multi-dimensional data and waveform characteristic data into the fusing prediction model, and predict the fusing probability of the bonding wire; The fuse protection module is used to analyze the fuse factors based on the fuse probability of the bonding wire and to perform fuse protection on the bonding wire through adaptive adjustment and alarm mechanism; The analysis report generation module is used to generate a fusing capacity analysis report and optimize the fusing prediction model after completing the fusing prediction and protection measures.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for analyzing the fusing capacity of a high-current bonding gold wire according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for analyzing the fusing capacity of a high-current bonding gold wire according to any one of claims 1 to 7 are implemented.

Images