FPC connector quality detection method and system based on multi-mode fusion

Abstract

The invention relates to the technical field of electronic equipment detection, and discloses an FPC connector quality detection method and system based on multi-modal fusion, and the method comprises the steps: obtaining an original multi-modal data set containing electrical stress, thermal stress and mechanical stress, and carrying out the filtering and time sequence alignment preprocessing, and obtaining a standardized multi-physical field sequence; constructing a three-dimensional space field intensity distribution diagram according to the sequence, and identifying field intensity anomalies by adopting density clustering so as to accurately position potential defects; learning a time sequence evolution mode based on the multi-moment observation value of the defect position to predict the future diffusion trend of the defect; and integrating the diffusion trend, the real-time field intensity and the environment variable to carry out comprehensive modeling and evaluation, and generating a dynamically updated and final quality detection report. According to the method, root cause diagnosis of connector defects, accurate three-dimensional space fault positioning and accurate prediction of future evolution trends can be realized, and the detection precision and predictability are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electronic equipment state monitoring, and in particular to a FPC connector quality detection method and system based on multi-modal fusion. BACKGROUND

[0002] In the field of electronic manufacturing and system integration, connector quality detection is a core link to ensure long-term stable operation of equipment and signal integrity. With the increasing demand for equipment reliability in the communication, automotive electronics and aerospace industries, developing high-precision and high-forecasting connector defect detection technology has become a key technical support for the development of high-end manufacturing industry.

[0003] In the prior art, connector quality detection relies on single physical dimension measurement methods, such as analyzing time domain reflection (TDR) and other electrical parameters to determine the pass state, or using X-ray imaging technology to obtain static two-dimensional structure information. Although this method can identify obvious defects to some extent, it has fundamental limitations. On the one hand, they separate the internal correlation of electrical, thermal and mechanical stress and other physical fields, and cannot capture the real causes and dynamic evolution process of defects under the interaction of multiple fields. More deeply, the existing technology lacks real-time insight into the dynamic changes of the internal field strength distribution of the connector, and fails to establish a correlation mechanism between the physical form of the defect (such as the three-dimensional structure of micro-cracks) and its evolution trend (such as the diffusion rate and direction of cracks) under the interaction of multiple physical fields, resulting in the inability to fundamentally accurately assess and predict the potential risks of defects.

[0004] In summary, the existing connector quality detection method cannot integrate multi-physical field information to diagnose the root cause of defects and predict the evolution trend, and has the technical problems of insufficient detection accuracy and lack of predictability. SUMMARY

[0005] The present application provides a FPC connector quality detection method and system based on multi-modal fusion to improve the accuracy of defect diagnosis and the reliability of evolution trend prediction in complex multi-physical field environment.

[0006] In a first aspect, to solve the above technical problems, the present application provides a FPC connector quality detection method based on multi-modal fusion, comprising: obtaining an original multi-modal data set inside the connector, the original multi-modal data set comprising electrical stress, thermal stress and mechanical stress data; performing mean filtering and time sequence alignment on the original multi-modal data set to obtain a standardized multi-physical field sequence; performing spatial feature extraction and time sequence construction on the standardized multi-physical field sequence to obtain a spatial field strength distribution map; If the field strength distribution in the spatial field strength distribution map exceeds the preset field strength threshold, density clustering and spatial correlation analysis are performed to obtain a potential defect area list; Spatial data of the potential defect area list is acquired, a connector internal three-dimensional structure model is constructed, structure analysis and abnormal point positioning are performed, and accurate defect position coordinates are obtained; Multi-time observation values of the accurate defect position coordinates are acquired, time series evolution mode learning and future trend prediction are performed on the observation values, and a future evolution path of the defect is obtained; The gradient change of the spatial field strength distribution map is calculated, and comprehensive modeling is performed in combination with the future evolution path to obtain a dynamic field strength update sequence; According to the dynamic field strength update sequence, in combination with the real-time acquired field strength distribution data, diffusion rate calculation and quality assessment are performed to obtain a final quality detection report.

[0007] Preferably, the mean filtering and time series alignment of the original multi-modal data set to obtain a standardized multi-physical field sequence comprises: The original multi-modal data set is smoothed by a mean filtering method to obtain a denoised data set; The timestamps of signals of different modalities in the denoised data set are synchronized, and missing values in the synchronization process are filled by a linear interpolation method to obtain a time series synchronized data set; The time series synchronized data set is normalized to map data of different physical dimensions to a unified scale to obtain a standardized multi-physical field sequence.

[0008] Preferably, the spatial feature extraction and time series construction of the standardized multi-physical field sequence to obtain a spatial field strength distribution map comprises: The standardized multi-physical field sequence is processed by a convolutional neural network, and multi-dimensional spatial features are extracted to generate a preliminary field strength distribution map; The field strength deviation value of the preliminary field strength distribution map is calculated, and the parameters of the convolutional neural network are optimized by a gradient descent method according to the field strength deviation value to obtain an optimized field strength distribution map; The optimized field strength distribution map is clustered and grouped, the spatial correlation between regions of each group is calculated, and the spatial correlation is dimensionally reduced by a principal component analysis method to obtain a reduced feature set; The reduced feature set is stored in a pre-established distributed database and organized by a time series index to obtain a spatial field strength distribution map.

[0009] Preferably, if the field strength distribution in the spatial field strength distribution map exceeds the preset field strength threshold, density clustering and spatial correlation analysis are performed to obtain a potential defect area list, including: The spatial field strength distribution map is subjected to median filtering preprocessing, and it is judged whether the data value after preprocessing exceeds the preset field strength threshold. The area exceeding the threshold is determined as an abnormal area; Density clustering is used to divide the abnormal area, and the area with high field strength value and dense space is grouped to obtain a high-density scene area set; The feature vector of each area in the high-density scene area set is extracted, and the spatial correlation between areas is calculated using the cosine similarity method to obtain a correlation matrix; The multi-dimensional features in the correlation matrix are reduced using linear discriminant analysis method to obtain a potential defect area list.

[0010] Preferably, the spatial data of the potential defect area list is obtained, a three-dimensional structure model of the connector is constructed, structure analysis and abnormal point positioning are performed, and accurate defect position coordinates are obtained, including: According to the potential defect area list, a stereoscopic microscopic three-dimensional scanning is performed on the target area to generate three-dimensional point cloud data; The three-dimensional point cloud data is subjected to data fusion processing, and multi-modal data such as field strength distribution and temperature signal are integrated to construct a preliminary three-dimensional structure model of the connector; The preliminary three-dimensional structure model is subjected to structure analysis and feature extraction, key structure features are identified, and geometric abnormal points are determined; The geometric abnormal points are subjected to coordinate calculation and correction to obtain accurate defect position coordinates.

[0011] Preferably, the multi-time observation value of the accurate defect position coordinates is obtained, the observation value is subjected to time series evolution pattern learning and future trend prediction to obtain the future evolution path of the defect, including: The multi-time observation value is subjected to timestamp sorting to obtain an ordered sequence of observation values, the difference value of adjacent observation values in the ordered sequence is calculated, and all difference values exceeding the preset change threshold are collected to form a defect change increment sequence; The defect change increment sequence is input into a time series model for layer-by-layer propagation and learning to obtain a state hidden representation representing the overall evolution pattern of the sequence; According to the state hidden representation, the trend continuation direction is judged and extrapolated to obtain the future evolution path of the defect.

[0012] Preferably, the gradient change of the spatial field strength distribution map is calculated, and the future evolution path is combined for comprehensive modeling to obtain a dynamic field strength update sequence, including: Gradient calculation is performed on the spatial field intensity distribution map to extract the rate of change of field intensity in space, and a gradient value sequence is obtained; It is judged whether the difference between adjacent values of the gradient value sequence exceeds a preset gradient stability threshold. When the threshold is exceeded, multiple possible branches of the future evolution path are fused to obtain a fused path branch sequence; According to the fused path branch sequence, the thermal stress concentration area of the non-uniform area is located, and the fluctuation sequence of the real-time environmental variable is determined; Data fusion is performed on the gradient value sequence, the position data of the thermal stress concentration area, and the fluctuation sequence of the environmental variable to finally generate the dynamic field intensity update sequence.

[0013] Preferably, according to the dynamic field intensity update sequence, combined with the real-time acquired field intensity distribution data, diffusion rate calculation and quality assessment are performed to obtain a final quality detection report, including: The dynamic field intensity update sequence is applied to the real-time acquired field intensity distribution data to generate predictive enhanced field intensity distribution data; Spatial feature extraction is performed on the predictive enhanced field intensity distribution data, and the change of the feature with time is calculated to obtain a defect diffusion rate vector; According to the defect diffusion rate vector, a three-dimensional model is built, and environmental fluctuation and thermal stress data are fused to generate an enhanced three-dimensional coordinate sequence; The enhanced three-dimensional coordinate sequence and the defect diffusion rate vector are compared with a preset quality standard to calculate the specific values of the geometric size and diffusion rate that exceed the standard limit, and finally a quality detection report is generated.

[0014] In a second aspect, the present application provides a FPC connector quality detection system based on multi-modal fusion, comprising: A multi-modal data acquisition module is used to acquire an original multi-modal data set inside the connector, and the original multi-modal data set includes electrical stress, thermal stress and mechanical stress data; A data preprocessing module is used to perform mean filtering and time sequence alignment on the original multi-modal data set to obtain a standardized multi-physical field sequence; A field intensity map construction module is used to perform spatial feature extraction and time sequence construction on the standardized multi-physical field sequence to obtain a spatial field intensity distribution map; A defect detection module is used to perform density clustering and spatial correlation analysis if the field intensity distribution in the spatial field intensity distribution map exceeds a preset field intensity threshold to obtain a potential defect area list; A three-dimensional defect positioning module is used to acquire spatial data of the potential defect area list, construct a three-dimensional structure model inside the connector, perform structure analysis and abnormal point positioning, and obtain accurate defect position coordinates; an evolution path prediction module configured to obtain multi-time observation values of the precise defect position coordinates, learn a time sequence evolution mode from the observation values, and predict a future evolution path of the defect; a dynamic field intensity update module configured to calculate gradient changes of the spatial field intensity distribution map, comprehensively model the future evolution path, and obtain a dynamic field intensity update sequence; a quality assessment and reporting module configured to calculate a diffusion rate and assess quality according to the dynamic field intensity update sequence and real-time acquired field intensity distribution data, and obtain a final quality detection report.

[0015] Compared with the prior art, the present application has the following beneficial effects: (1) The present application obtains multi-modal data such as electrical, thermal and mechanical stress inside the connector, and constructs a unified spatial field intensity distribution map through time sequence alignment and fusion processing. This method overcomes the limitations of the prior art of analyzing multiple physical fields separately, can reveal the real causes of defects under the action of multiple field coupling, and thus realizes comprehensive and deep diagnosis of the internal state of the connector, fundamentally improving the identification accuracy of early potential defects.

[0016] (2) After identifying the field intensity abnormal area, the present application further constructs a connector internal structure model through a three-dimensional reconstruction algorithm to realize accurate three-dimensional coordinate positioning of the defect. This positioning mechanism from macro field intensity distribution to micro three-dimensional structure accurately associates abstract performance abnormalities with specific physical forms, solves the problem of the prior art of only providing two-dimensional or fuzzy position information, not only realizes visualization of the defect, but also provides accurate geometric input for subsequent evolution prediction.

[0017] (3) The present application collects multi-time observation values of the positioned defect coordinates to form a time sequence, and predicts the future evolution path of the defect by using time sequence evolution mode learning. This converts the traditional static "snapshot" detection into dynamic "process" analysis, learns the historical law of defect development, gives the detection method the scientific inference ability of future trends, effectively solves the core pain point of the prior art of lacking predictability, and upgrades quality assessment from post-diagnosis to pre-warning.

[0018] (4) The present application comprehensively models the predicted future evolution path and the gradient changes of the spatial field intensity to generate a dynamic field intensity update sequence, and calculates the defect diffusion rate in combination with real-time data to complete quality assessment. This method creatively establishes a closed-loop correlation between defect physical form evolution and dynamic response of multiple physical fields, makes the final quality assessment report have foresight, can quantify the future risk level of the defect, and provides strong technical support for ensuring long-term reliability of the system. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a flowchart of a FPC connector quality detection method based on multi-modal fusion provided by the first embodiment of the present application; Figure 2 is a structural diagram of a FPC connector quality detection system based on multi-modal fusion provided by the second embodiment of the present application. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0021] With reference to Figure 1 The first embodiment of the present application provides a FPC connector quality detection method based on multi-modal fusion, comprising the following steps: S11, obtaining an original multi-modal data set inside the connector, wherein the original multi-modal data set comprises electrical stress, thermal stress and mechanical stress data; S12, performing mean filtering and time sequence alignment on the original multi-modal data set to obtain a standardized multi-physical field sequence; S13, performing spatial feature extraction and time sequence construction on the standardized multi-physical field sequence to obtain a spatial field intensity distribution map; S14, if the field intensity distribution in the spatial field intensity distribution map exceeds a preset field intensity threshold, performing density clustering and spatial correlation analysis to obtain a potential defect area list; S15, obtaining spatial data of the potential defect area list, constructing a three-dimensional structure model inside the connector, performing structure analysis and abnormal point positioning to obtain accurate defect position coordinates; S16, obtaining multi-time observation values of the accurate defect position coordinates, performing time sequence evolution mode learning and future trend prediction on the observation values to obtain a future evolution path of the defect; S17, calculating the gradient change of the spatial field intensity distribution map, combining the future evolution path to perform comprehensive modeling to obtain a dynamic field intensity update sequence; S18, according to the dynamic field intensity update sequence, combining the real-time obtained field intensity distribution data to perform diffusion rate calculation and quality evaluation to obtain a final quality detection report.

[0022] In step S11, an original multi-modal data set inside the connector is obtained, wherein the original multi-modal data set comprises electrical stress, thermal stress and mechanical stress data.

[0023] It should be noted that the core of step S11 is to synchronously and highly accurately collect data of multiple physical dimensions of key nodes in the FPC connector through an integrated multi-sensor array. The array is not a simple collection of sensors, but a system that works cooperatively, aiming to capture weak signal responses in different physical fields that are coupled to each other caused by potential defects, thereby constructing an original multi-modal data set that can comprehensively characterize the health status of the connector.

[0024] Specifically, the acquisition of the original multi-modal data set includes the following contents: The acquisition of electrical stress data is achieved by deploying a non-contact Hall Effect Current Sensor near the key conductive path of the connector. According to the Hall Effect principle, the sensor accurately calculates the amount of current flowing by measuring the change in the magnetic field strength around the conductor. For example, the current is continuously monitored at a sampling frequency of 10 kHz to generate a current time series reflecting load fluctuations, instantaneous overcurrent, or contact resistance changes, with the quantitative unit being amperes (A).

[0025] The acquisition of thermal stress data is achieved by attaching a miniature K-type Thermocouple to hot spots such as the connector plug interface and crimped terminals. The thermocouple uses the Seebeck effect to convert the weak voltage generated by the temperature difference at the contact points of different metals into accurate temperature readings. For example, temperature collection is performed at a sampling frequency of 20 Hz to generate a temperature time series reflecting Joule heat accumulation, poor heat dissipation, or environmental temperature effects, with the quantitative unit being degrees Celsius.

[0026] The acquisition of mechanical stress data is achieved by attaching a resistive strain gauge to the stress concentration area of the connector housing or FPC ribbon root. The core principle of the strain gauge is the piezoresistive effect of metal, and its resistance value will change linearly with the physical deformation (stretching or compression) of the base material. For example, the change in resistance value is measured by a high-precision bridge circuit at a sampling frequency of 5 kHz, and converted into a dimensionless micro-strain time series to represent the structural response of the connector under vibration, impact, or plug-in operation.

[0027] In step S12, the original multi-modal data set is subjected to mean filtering and time series alignment to obtain a standardized multi-physical field sequence. It should be noted that in the first embodiment of the present application, this processing step specifically includes steps S121 to S123: S121, the original multi-modal data set is smoothed by a mean filtering method to obtain a denoised data set; S122, time stamp synchronization processing is performed on signals of different modalities in the denoised data set, and a linear interpolation method is used to fill in missing values in the synchronization process, to obtain a time sequence synchronized data set; S123, the time sequence synchronized data set is normalized to map data of different physical dimensions to a unified scale, to obtain a standardized multi-physical field sequence.

[0028] In step S121, the original multi-modal data set is smoothed by using a mean filter method to obtain a denoised data set. Specifically, this step aims to suppress high-frequency noise and transient interference introduced in the acquisition process. In this embodiment, a moving average filter is used to achieve this function. The core principle of the filter is to take a certain point in the data sequence as the center, take the N data points before and after it to form a sliding window, calculate the arithmetic mean of all data points in the window, and take the average as the new value of the center point after filtering. For example, the size N of the sliding window can be set to 5 sampling points. It should be noted that the value of the window size N is determined based on experimental statistical analysis on historical noise data sets. By testing the smoothing effect of different window values (for example, N=3, 5, 7) on typical noise signals, and evaluating the waveform fidelity of the original effective signal (such as electrical stress peaks), an optimal value (for example, N=5) is finally selected, which can maximize the signal-to-noise ratio while controlling the signal distortion within a predetermined range. When there is an abnormal peak in the electrical stress data due to electromagnetic crosstalk, this processing can effectively smooth the peak while preserving the low-frequency trend characteristics of the signal.

[0029] It should be noted that the start of the filtering process can be triggered by a preset noise level threshold. The threshold is set by statistical analysis of noise data of the same type of connector under typical working conditions, aiming to balance the denoising effect and signal fidelity. For example, when the signal-to-noise ratio (SNR) of the signal is lower than 15 decibels, mean filtering is started. In a preferred scheme, if the noise level after mean filtering is still higher than a second threshold (for example, the signal-to-noise ratio is still lower than 10 decibels), a median filter (Median Filter) can be further enabled for secondary processing to deal with stronger pulse noise.

[0030] In step S122, the signals of different modes in the denoised dataset are time-stamped and synchronized, and missing values ​​in the synchronization process are filled in using a linear interpolation method to obtain a time-synchronized dataset. Specifically, this step aims to solve the problem of inconsistent sampling frequencies caused by the physical characteristics of different sensors. This embodiment uses resampling technology to unify the time series data of all modes onto a common time reference grid. The frequency of this common grid is set according to the Nyquist-Shannon sampling theorem, and its frequency should be at least twice the highest effective frequency of all input signals. For example, if the mechanical stress signal contains vibration information up to 100 Hz, the frequency of the common time grid can be set to 250 Hz.

[0031] During resampling, for sequences with low original sampling rates (e.g., a 10 Hz temperature sequence), linear interpolation is used to estimate their values ​​at the missing time points in the new time grid. This algorithm works by creating a linear function between two known, adjacent data points and calculating the function value at any intermediate point. The calculation process is as follows: First, locate the anchor point: for any target time point on the unified time grid that needs interpolation. First, in the original low-frequency time series, locate the two nearest known real data points that are exactly before and after it. and ,in .

[0032] Then, calculate the rate of change: calculate the linear rate of change (i.e., the slope) between these two known "anchor points". The calculation formula is as follows: .

[0033] Finally, interpolation is calculated: Based on the point-slope linear equation, the target time point is calculated. Corresponding interpolation The calculation formula is as follows: By repeating the above process for all missing points, a complete low-frequency data stream that is perfectly time-aligned with the high-frequency data stream can be generated.

[0034] In step S123, the time-synchronized dataset is normalized to map data of different physical dimensions to a uniform scale, obtaining a standardized multi-physical field sequence. Specifically, this step aims to eliminate the scale difference between different physical dimensions (such as amperes for current, degrees Celsius for temperature, and micro-strain for strain), avoiding the dominance or neglect of model parameter updates due to the large or small value range of features in subsequent machine learning model training. The present embodiment adopts the Min-Max Normalization method. The core principle of this method is to strictly map the original data of any interval to the interval through linear transformation. Its transformation function is: ; where X is the original data point, and are the minimum and maximum values of the data sequence in a period of time. For example, a temperature data range of [25℃, 85℃], a reading of 65℃ will be normalized to (65-25) / (85-25)≈0.67. It should be noted that mapping data to the interval [0, 1] is a well-known technical practice in the field of deep learning, which can effectively improve the convergence speed and stability of models based on gradient descent algorithm.

[0035] In step S13, the standardized multi-physical field sequence is subjected to spatial feature extraction and time-series construction, obtaining a spatial field intensity distribution map. It should be noted that in the first embodiment of the present application, this processing step specifically includes steps S131 to S134: S131, the standardized multi-physical field sequence is processed using a convolutional neural network, and multi-dimensional spatial features are extracted to generate a preliminary field intensity distribution map; S132, the field intensity deviation value of the preliminary field intensity distribution map is calculated, and the parameters of the convolutional neural network are optimized using the gradient descent method according to the field intensity deviation value, obtaining an optimized field intensity distribution map; S133, the optimized field intensity distribution map is clustered and grouped, the spatial correlation between each group region is calculated, and the principal component analysis method is used to reduce the dimension of the spatial correlation, obtaining a reduced feature set; S134, the reduced feature set is stored in a pre-established distributed database, and is organized using time series index, obtaining a spatial field intensity distribution map.

[0036] In step S131, the standardized multi-physical field sequence is processed using a convolutional neural network to extract multi-dimensional spatial features and generate a preliminary field intensity distribution map. Specifically, the standardized multi-physical field sequence is first constructed into a three-dimensional tensor with dimensions such as [N, M, 3], where N x M represents the spatial grid resolution of the connector internal monitoring area, and 3 represents the three physical field channels of electricity, heat, and mechanics. The convolutional neural network (CNN) used in this embodiment is a hybrid structure containing two convolutional layers and two fully connected layers. The convolutional layer of the network uses a 3 x 3 convolution kernel (the number of convolution kernels can be 16-64) and a rectified linear unit (ReLU) as the activation function. The core principle is to use the convolution kernel as a learnable feature filter to automatically extract local spatial features such as field intensity boundaries and gradient discontinuities by sliding in space. The fully connected layer is responsible for nonlinearly combining the extracted local features, and finally outputs a single-channel preliminary field intensity distribution map with the same size as the input grid N x M.

[0037] It should be noted that the CNN model is pre-trained through supervised learning. The training data set comes from simulation data obtained by finite element analysis (FEA) of the same type of FPC connector under various working conditions. The simulation input (multi-physical field excitation) is used as the model input, and the known field intensity real distribution map obtained by simulation is used as the label (Ground Truth). The mean squared error (MSE) is used as the loss function, and the Adam optimizer is used for iterative training. The learning rate is usually set between 0.001 and 0.0001, and the model converges until the model converges.

[0038] In step S132, the field strength deviation value of the preliminary field strength distribution map is calculated, and the parameters of the convolutional neural network are optimized according to the field strength deviation value by using a gradient descent method to obtain an optimized field strength distribution map. Specifically, in order to make the pre-trained model better adapt to the real-time working condition of the current specific connector, online optimization is required. First, the mean square error (MSE) between the preliminary field strength distribution map and the sparse real sensor readings is calculated as the field strength deviation value. Then, if the deviation value exceeds a preset "model optimization trigger threshold" (which can be determined by cross-validation, ROC curve analysis or historical fault data statistics, and is usually set in the MSE range of 0.1-0.3, which can be adjusted according to the model accuracy requirement), the gradient descent (Gradient Descent) algorithm is started. The core principle of this algorithm is to adjust the weight parameters of the CNN model (especially the fully connected layer) in the direction of the fastest gradient descent of the loss function with a preset learning rate (learning rate, for example 0.001).

[0039] In step S133, the optimized field strength distribution map is clustered and grouped, the spatial correlation between each grouped region is calculated, and the principal component analysis method is used to reduce the dimension of the spatial correlation to obtain a reduced feature set. Specifically, first, the K-means clustering (K-Means Clustering) algorithm is used to divide the region of the optimized field strength distribution map. This algorithm divides all pixel points in the map into a preset K clusters according to their field strength values through iterative calculation, so that the variance of points within the cluster is minimized. For example, by setting K=3, the field strength map can be automatically divided into three regions of high, medium and low levels. Then, the Euclidean distance between the centroids of each region is calculated to construct a relationship matrix representing the spatial proximity relationship of each field strength region.

[0040] Finally, the relationship matrix is subjected to principal component analysis (Principal Component Analysis, PCA). The core principle of PCA is to project high-dimensional correlated feature vectors onto a set of low-dimensional orthogonal coordinate systems through linear transformation, and these new coordinate axes are called principal components. This embodiment retains the first k principal components (for example, k=3) that can explain more than 90% of the cumulative variance contribution rate of the original data. It should be noted that the threshold of 90% is an engineering experience value that balances the completeness of information retention and data compression efficiency, and those skilled in the art can adjust it within the range of 85% to 98% according to the specific accuracy requirement.

[0041] In step S134, the reduced dimension feature set is stored in a pre-established distributed database and organized using time series indexing to obtain a spatial field intensity distribution map. Specifically, to meet the efficient storage and query requirements of massive time series data, the present embodiment uses a distributed database based on time series (for example, InfluxDB). The database uses sharding technology to horizontally expand data to multiple physical nodes according to time range (for example, one shard per 24 hours). At the same time, an efficient index structure such as B+ tree is established for the data primary key (timestamp). The core advantage of this architecture is that when performing a time range query (for example, retrieving all field intensity features in the past hour), the database engine can directly locate the corresponding data shard and index location, avoiding full table scanning, and thus achieving millisecond-level query response. Finally, this three-dimensional feature dataset organized by timestamp and efficiently queryable constitutes a dynamic, traceable spatial field intensity distribution map sequence.

[0042] In step S14, if the field intensity distribution in the spatial field intensity distribution map exceeds a preset field intensity threshold, density clustering and spatial correlation analysis are performed to obtain a potential defect area list. It should be noted that in the first embodiment of the present application, this processing step specifically includes steps S141 to S144: S141, median filtering preprocessing is performed on the spatial field intensity distribution map, and it is determined whether the preprocessed data value exceeds a preset field intensity threshold. The area exceeding the threshold is determined as an abnormal area; S142, density clustering is used to divide the abnormal area, and the areas with high field intensity values and dense in space are grouped to obtain a high-density scene area set; S143, the feature vectors of each area in the high-density scene area set are extracted, and the spatial correlation between areas is calculated using the cosine similarity method to obtain a correlation matrix; S144, linear discriminant analysis method is used to reduce the dimension of the multi-dimensional features in the correlation matrix to obtain a potential defect area list.

[0043] In step S141, the spatial field intensity distribution map is median filtered and preprocessed, and it is determined whether the preprocessed data value exceeds a preset field intensity threshold value, and the area exceeding the threshold value is determined as an abnormal area. Specifically, first, the input field intensity distribution map is median filtered, and the filter replaces the value of each data point with the median value of the numerical values in its neighborhood window, effectively eliminating isolated noise points. For example, for the sequence [10, 12, 50, 11, 13], the median value 11 will be used to replace the abnormal value 50. Subsequently, the numerical value of each point in the filtered map is compared with a preset defect activation threshold value. The threshold value (for example, 15V / m) is set based on statistical analysis of the field intensity data of historical connectors in the normal aging process, and the 99% percentile of the distribution is selected, so as to ensure that only statistically significant abnormalities are marked.

[0044] In step S142, the abnormal areas are divided by using density clustering, the areas with high field intensity values and high spatial density are grouped, and a high-density scene area set is obtained. Specifically, the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm is used in this embodiment. The core principle of DBSCAN is not based on the distance to the cluster center, but on the "density reachability" of the area. It defines the neighborhood by two key parameters: neighborhood radius and minimum neighborhood sample number MinPts of core points. The algorithm starts from an unvisited core point, and recursively merges all density-reachable points of the core point into a cluster. For example, the neighborhood radius is set to 1.5mm, and the minimum sample number MinPts is set to 10 data points. The values of these two parameters are based on the physical design specifications of the connector and the spatial resolution of the sensor, and correspond to a minimum defect size that is considered meaningful in a physical sense. DBSCAN can discover defect clusters of any shape and automatically identify sparse, density-unqualified abnormal points as noise, which is very suitable for irregular defect identification.

[0045] In step S143, the feature vectors of the areas in the high-density scene area set are extracted, and the spatial correlation between the areas is calculated by using the cosine similarity method, and a correlation matrix is obtained. Specifically, first, a multi-dimensional feature vector is constructed for each high-density area (i.e. defect cluster) obtained in S142. The vector contains quantitative indicators describing the characteristics of the area, for example, the vector can be [average field intensity value, peak temperature, area volume, geometric irregularity]. Subsequently, the cosine similarity (Cosine Similarity) is used to calculate the feature vectors and The principle is to calculate the cosine value of the angle between two vectors in a multi-dimensional space: ; The result of this value ranges between [-1, 1], and the closer the value is to 1, the more similar the intrinsic physical characteristics of the two defect clusters are, and they are likely to be caused by the same root cause (e.g., the extension of the same crack). For example, the feature vectors of two regions are [18, 10, 5] and [20, 12, 6] respectively, and the calculated cosine similarity can be as high as 0.98.

[0046] In step S144, the multi-dimensional features in the correlation matrix are processed by dimension reduction using a linear discriminant analysis method to obtain a list of potential defect regions. Specifically, the correlation matrix generated in S143 and the feature vectors of each region are combined to form a high-dimensional feature space. Linear Discriminant Analysis (LDA) is used to reduce the dimension of this space. LDA is a supervised dimension reduction technique, and its core principle is to find a projection direction that maximizes the distance between samples of different classes (e.g., "high-risk cracks" and "general hot spots" labeled according to historical data) and minimizes the variance of samples within the same class. Through this method, high-dimensional features (e.g., more than 10 feature dimensions) can be projected into a low-dimensional space (e.g., 2D) that best reflects the risk level of defects, and finally a clear, risk level ordered list of potential defect regions is generated. The linear discriminant analysis can be used according to whether there is labeled data in actual application, if there is labeled information of historical defect types, using LDA can effectively extract discriminative features; if there is no label, it can be replaced by Principal Component Analysis (PCA) or other unsupervised dimension reduction methods to achieve feature compression and visualization.

[0047] In step S15, the spatial data of the list of potential defect regions is obtained, a three-dimensional structure model of the connector is constructed, structure analysis and abnormal point positioning are performed, and accurate defect position coordinates are obtained. It should be noted that in the first embodiment of the present application, this processing step specifically includes steps S151 to S154: S151, according to the list of potential defect regions, performing stereoscopic microscopic three-dimensional scanning on the target region to generate three-dimensional point cloud data; S152, performing data fusion processing on the three-dimensional point cloud data, integrating multi-modal data such as field intensity distribution and temperature signal, and constructing a preliminary three-dimensional structure model of the connector; S153, performing structure analysis and feature extraction on the preliminary three-dimensional structure model, identifying key structural features, and determining geometric abnormal points therein; S154, coordinate calculation and correction are performed on the geometric abnormal points to obtain accurate defect position coordinates.

[0048] In step S151, stereomicroscopic three-dimensional scanning is performed on the target area according to the list of potential defect areas to generate three-dimensional point cloud data. Specifically, the core of this step is to obtain a high-fidelity three-dimensional geometric morphology of the defect area. This embodiment adopts a scanning technology based on the principle of binocular stereo vision. This technology simulates human visual parallax to synchronously collect two-dimensional images of the target area from two or more different angles, and reconstructs the three-dimensional spatial coordinates of each pixel point by calculating the pixel displacement (parallax) of the homonymous points in different images using the triangulation principle. For example, for a welding spot defect area in the list, the scanning system can perform scanning at a spatial resolution of 0.1 mm, and finally generate a three-dimensional point cloud data set composed of tens of thousands of (x, y, z) coordinate points.

[0049] In step S152, data fusion processing is performed on the three-dimensional point cloud data to integrate multi-modal data such as field intensity distribution and temperature signals, and to construct a preliminary three-dimensional structure model of the connector. Specifically, this step aims to accurately register and fuse the pure geometric model (target point cloud) obtained in S151 with the physical field data (source point cloud) with spatial coordinates obtained in the previous steps. This embodiment uses the Iterative Closest Point (ICP) algorithm to achieve accurate spatial alignment.

[0050] The iteration process of the ICP algorithm is as follows: first, point pair association, for each point in the source point cloud, find the closest point in the target point cloud to form an initial matching point pair. Then, transformation solving, construct a loss function aimed at minimizing the mean square distance error between all matching point pairs, and solve the optimal rigid transformation parameters, i.e. a rotation matrix R and a translation vector t, that minimize the loss function by numerical optimization methods such as Singular Value Decomposition (SVD). Next, point cloud transformation, apply the rotation matrix R and translation vector t obtained to the entire source point cloud. Finally, iteration termination judgment, calculate the mean square distance error of the transformed source point cloud and the target point cloud, if the error is less than the preset convergence threshold or the iteration number reaches the upper limit (e.g. 100 times), the algorithm terminates and outputs the final transformation matrix.

[0051] After the two point cloud coordinate systems are aligned by the ICP algorithm, the multi-modal data is weighted and fused. For example, the comprehensive attribute of any point p in the model can be calculated by the following formula: ; wherein are the electrical, thermal, mechanical stress values of the point, are the corresponding weight coefficients (for example, , is the sum of the weights). It should be noted that the weight coefficients are set on the basis of statistical analysis of historical failure samples or automatically learned through a machine learning model (such as a neural network), aiming to highlight the physical field information most strongly associated with this type of defect, rather than being fixed.

[0052] In step S153, the preliminary three-dimensional structure model is subjected to structure analysis and feature extraction, key structural features are identified, and geometric abnormal points are determined. Specifically, this step automatically "understands" the geometric shape of the three-dimensional model through an algorithm. In this embodiment, curvature analysis based on point cloud normal vectors is used to identify geometric abnormalities. By calculating the local curvature of each point on the model surface, the mutation region of the geometric shape can be effectively identified. The setting of this curvature threshold is based on statistical analysis of the geometric features of good samples, and the upper limit of the normal range is selected, for example, the 99.5% quantile. When the calculated curvature of a point in the model exceeds this threshold, the point is marked as a geometric abnormal point. For example, through this method, the curvature abnormal region caused by micro-cracks can be accurately identified from the surface of a 2.5 mm diameter solder joint.

[0053] In step S154, the coordinates of the geometric abnormal points are calculated and corrected to obtain the precise defect position coordinates. Specifically, this step aims to finally calibrate the coordinates of the abnormal points identified in S153. In this embodiment, at least three fiducial markers pre-set on the connector are introduced to perform global coordinate transformation and optimization with the initial coordinates of the abnormal points obtained in S151. This process calculates the optimal estimated coordinates of the abnormal points by minimizing the re-projection error of all points in three-dimensional space. For example, an abnormal point with an initial coordinate of (x = 15.2 mm, y = 8.7 mm, z = 3.1 mm) is corrected by the fiducial markers, and the final output of the precise defect position coordinates is (x = 15.02 mm, y = 8.75 mm, z = 3.11 mm).

[0054] In step S16, the multi-time observation values of the precise defect position coordinates are obtained, the observation values are subjected to time series evolution pattern learning and future trend prediction, and the future evolution path of the defect is obtained. It should be noted that in the first embodiment of the present application, this processing step specifically includes steps S161 to S163: S161, time stamp sorting is performed on the multi-time observation values to obtain an ordered sequence of observation values, difference values between adjacent observation values in the ordered sequence are calculated, and all difference values exceeding a preset change threshold are collected to form a defect change increment sequence; S162, the defect change increment sequence is input to a time sequence model for layer-by-layer propagation and learning to obtain a state hidden representation representing the overall evolution pattern of the sequence; S163, a trend continuation direction is determined according to the state hidden representation, and extrapolation is performed to obtain a future evolution path of the defect.

[0055] In step S161, the multi-time observation values are time stamp sorted to obtain an ordered sequence of observation values, difference values between adjacent observation values in the ordered sequence are calculated, and all difference values exceeding a preset change threshold are collected to form a defect change increment sequence. Specifically, this step aims to convert the original, absolute observation value sequence into a sparse, incremental sequence that better reflects dynamic changes. First, the multi-time observation values (such as the curvature of defect points) collected within, for example, 24 consecutive hours are arranged in ascending order of time stamp. Then, the difference between each two adjacent observation values in the ordered sequence is calculated, and the absolute value of the difference is compared with a preset "evolution significance threshold". The threshold (for example, a curvature change of 0.02 mm) is set based on material fatigue theory and experimental statistics or by observing the minimum significant change amount of defect expansion in accelerated life tests to ensure its engineering applicability and statistical significance, which represents the minimum change amount sufficient to exclude measurement noise and mark the real evolution of the physical structure. All difference values exceeding the threshold (retaining their signs) are collected to form the final defect change increment sequence.

[0056] In step S162, the defect change increment sequence is input into a time sequence model for layer-by-layer propagation and learning to obtain a state hidden representation representing the overall evolution mode of the sequence. Specifically, the time sequence model used in this embodiment is a stacked long short-term memory (LSTM). Specifically, the network is composed of an input layer, two stacked LSTM layers, and an output layer. The input layer receives the defect change increment sequence generated in S161. The first LSTM layer contains 64 hidden units, and the second LSTM layer contains 32 hidden units. The core of the LSTM unit is its internal "gating mechanism" (input gate, forget gate, output gate), which enables it to selectively remember long-term dependencies in the sequence (e.g., cumulative damage effect of defects) while forgetting unimportant short-term fluctuations. The final output layer outputs a fixed-dimensional vector, i.e., the state hidden representation. The LSTM model is pre-trained through supervised learning. The training data set is derived from a large amount of historical full-life monitoring data of similar connectors. The defect change increment sequence of the historical sample is used as the model input, and its known, true subsequent evolution path is used as the label. The root mean square error (RMSE) is used as the loss function, and the Adam optimizer is used for iterative training, with a learning rate usually set between 0.001 and 0.0001, until the model can accurately output a hidden state representing the future trend based on the input historical change pattern.

[0057] In step S163, the trend continuation direction is determined according to the state hidden representation and extrapolated to obtain the future evolution path of the defect. Specifically, the state hidden representation output by S162 is a highly condensed feature vector, and the direction and norm of the vector contain the cumulative trend of defect evolution. In this embodiment, a fully connected layer (also known as a prediction head) is used to decode the state hidden representation. The prediction head is trained to map the input hidden representation to a specific future path description. For example, if the dimensions representing "positive growth" and "acceleration" in the hidden representation vector have higher weight values, the decoder will output a prediction path of accelerated expansion, such as: "It is predicted that in the next 12 hours, the defect curvature will increase from 0.12 mm to 0.18 mm, and will mainly expand along the Z-axis direction".

[0058] In step S17, the gradient change of the spatial field strength distribution map is calculated, and a comprehensive model is established in combination with the future evolution path to obtain a dynamic field strength update sequence. It should be noted that in the first embodiment of the present application, this processing step specifically includes steps S171 to S174: S171, calculating the gradient of the spatial field strength distribution map to extract the rate of change of the field strength in space to obtain a gradient value sequence; S172, judge whether the difference between adjacent values of the gradient value sequence exceeds a preset gradient stability threshold, and when the threshold is exceeded, fuse multiple possible branches of the future evolution path to obtain a fused path branch sequence; S173, according to the fused path branch sequence, locate the thermal stress concentration area of the non-uniform region and determine the fluctuation sequence of the real-time environmental variable; S174, data fusion is performed on the gradient value sequence, the position data of the thermal stress concentration area and the fluctuation sequence of the environmental variable to finally generate the dynamic field strength update sequence.

[0059] In step S171, the gradient of the spatial field strength distribution map is calculated to extract the rate of change of the field strength in space, and a gradient value sequence is obtained. Specifically, this step is realized by a three-dimensional Sobel operator. The Sobel operator is a discrete differential operator, and its core principle is to approximate the partial derivative of each data point in three orthogonal directions in space by performing convolution operation on the field strength distribution map and a 3x3x3 convolution kernel. Finally, the gradient of each point is calculated as the norm of its gradient vector, for example, 0.22V / m / mm, which represents the degree of change of the field strength in space.

[0060] In step S172, it is judged whether the difference between adjacent values of the gradient value sequence exceeds a preset gradient stability threshold, and when the threshold is exceeded, multiple possible branches of the future evolution path are fused to obtain a fused path branch sequence. Specifically, the first-order difference of the gradient value time sequence obtained in S171 is first calculated. Then, the difference value is compared with the gradient stability threshold. The setting of this threshold (for example, 0.05V / m / mm) is based on the statistical analysis of the natural fluctuation of the field strength gradient of the connector in the stable working state for 100 hours of continuous operation, and is determined by selecting the 98% quantile point of the distribution. When the difference value exceeds this threshold, it indicates that the field strength change enters an unstable state, and at this time, path fusion is enabled. The fusion is realized by weighting and averaging the multiple predicted path branches output by S16 according to their respective confidence levels, and integrating them into a single fused path branch sequence with the highest expected value.

[0061] In step S173, the thermal stress concentration area of the non-uniform area is located according to the fusion path branch sequence, and the fluctuation sequence of the real-time environmental variable is determined. Specifically, this step processes two inputs in parallel: first, the fused defect propagation path is taken as input, and simulation calculation is performed in a pre-constructed connector thermal coupling model based on the finite element method (FEM) (the FEM model is constructed based on the material properties and structural design parameters of the connector, and is verified by experimental data such as thermal imaging), so as to locate the three-dimensional coordinates (for example, peak stress 2.5 MPa) of the new thermal stress concentration area caused by the path. Second, the real-time collected environmental variable (such as humidity, vibration) sequence is processed by a pre-trained LSTM network, the fluctuation mode is analyzed, and a state hidden representation vector (for example, [0.08, 0.12]) that can represent the short-term change trend is output.

[0062] In step S174, the gradient value sequence, the position data of the thermal stress concentration area, and the fluctuation sequence of the environmental variable are fused to finally generate the dynamic field strength update sequence. Specifically, this step performs final nonlinear fusion on multiple feature streams generated in the previous steps through a multi-layer perceptron (MLP) fusion network. The input layer of the MLP network receives a concatenated feature vector, which at least includes the gradient value at the current time, the three-dimensional coordinates of the thermal stress concentration area, and the state hidden representation of the environmental variable. After training, the output layer of the network can generate an accurate scalar value, which is the predicted update value of the field strength at the next time. For example, according to the fusion analysis, the field strength value of 4.8 V / m at the current time is updated to the predicted value of 4.6 V / m at the next time. By repeating this process for all key points in the field strength map, the final dynamic field strength update sequence is generated.

[0063] In step S18, according to the dynamic field strength update sequence, the diffusion rate calculation and quality assessment are performed in combination with the real-time acquired field strength distribution data to obtain the final quality detection report. It should be noted that in the first embodiment of the present application, this processing step specifically includes steps S181 to S184: S181, applying the dynamic field strength update sequence to the real-time acquired field strength distribution data to generate predictive enhanced field strength distribution data; S182, performing spatial feature extraction on the predictive enhanced field strength distribution data and calculating the change of the feature over time to obtain a defect diffusion rate vector; S183, performing three-dimensional modeling according to the defect diffusion rate vector and fusing environmental fluctuation and thermal stress data to generate an enhanced three-dimensional coordinate sequence; S184, comparing the enhanced three-dimensional coordinate sequence with the defect diffusion rate vector to a preset quality standard, calculating the specific values of the geometric size and diffusion rate that exceed the standard limits, and finally generating a quality detection report.

[0064] In step S181, the dynamic field strength update sequence is applied to the real-time acquired field strength distribution data to generate predictive enhanced field strength distribution data. Specifically, this step fuses prediction and observation through a Kalman Filter framework. The dynamic field strength update sequence output by S17 is used as the prediction step of the framework to provide a priori estimate of the current field strength state, while the real-time acquired field strength distribution data is used as the update step to provide a true measurement of the current state. The Kalman Filter optimally combines the prediction and measurement values according to their respective uncertainties (quantified by the process noise covariance and measurement noise covariance matrix) by calculating the Kalman Gain, and finally outputs a posteriori estimate, i.e. the predictive enhanced field strength distribution data. This data is closer to the true state than either pure prediction or pure measurement; the process noise covariance matrix can be initialized by covariance estimation of historical prediction errors; the measurement noise covariance matrix can be determined by sensor calibration data; in a preferred scheme, adaptive Kalman filtering (AKF) technology can be used for real-time adjustment.

[0065] In step S182, spatial feature extraction is performed on the predictive enhanced field strength distribution data, and the change of the feature over time is calculated to obtain a defect diffusion rate vector. Specifically, this step first uses an Isocontour Analysis algorithm to extract specific field strength isoclines representing the defect boundary on the enhanced field strength map generated in S181. Then, by comparing the changes in the position of the isocline centroid at two consecutive time steps (e.g. with a time interval of 100 milliseconds), the displacement vector of the isocline in space is calculated. The defect diffusion rate vector is obtained by dividing the displacement vector by the time interval.

[0066] In step S183, a three-dimensional model is constructed according to the defect diffusion rate vector, and environmental fluctuations and thermal stress data are fused to generate an enhanced three-dimensional coordinate sequence. Specifically, this step constructs a three-dimensional trajectory of the defect core region by time integrating the instantaneous rate vector obtained in S182. The initial point of the trajectory is derived from the accurate defect position coordinates determined in S15. To further improve the accuracy of the trajectory, the model also fuses the environmental fluctuation trend and thermal stress concentration zone position data obtained in S173 to fine-tune the trajectory, to compensate for the disturbance of external factors on the defect propagation path, and finally generate the enhanced three-dimensional coordinate sequence.

[0067] In step S184, the enhanced three-dimensional coordinate sequence is compared with the defect propagation rate vector and a preset quality standard, the specific values of the geometric size and the propagation rate exceeding the standard limit are calculated, and finally a quality detection report is generated. Specifically, the quality standard is a pre-defined multi-dimensional safety envelope. The setting of the envelope is based on the design specification of the connector, the material fatigue limit and the industry safety standard (for example, the GJB standard of the aviation industry). The evaluation process includes geometric evaluation and dynamic evaluation. In the geometric evaluation, it is judged whether the enhanced three-dimensional coordinate sequence generated in S183 invades the "prohibited entry" area defined in the safety envelope, and the specific value of the maximum size (for example, 0.5 millimeters) exceeding the allowed maximum size is calculated. In the dynamic evaluation, it is judged whether the modulus of the defect propagation rate vector obtained in S182 exceeds the "maximum allowed propagation rate" (for example, 0.09 millimeters / second) defined in the safety envelope, and the specific value of the excess rate is calculated. Finally, these quantified deviation values are written together with the corresponding risk level to generate the final quality detection report.

[0068] Referring to Figure 2 The second embodiment of the present application provides a multi-modal fusion-based FPC connector quality detection system, comprising: A multi-modal data acquisition module is configured to acquire an original multi-modal data set inside the connector, wherein the original multi-modal data set comprises electrical stress, thermal stress and mechanical stress data. A data preprocessing module is configured to perform mean filtering and time series alignment on the original multi-modal data set to obtain a standardized multi-physical field sequence. A field intensity map construction module is configured to perform spatial feature extraction and time series construction on the standardized multi-physical field sequence to obtain a spatial field intensity distribution map. A defect detection module is configured to perform density clustering and spatial correlation analysis if the field intensity distribution in the spatial field intensity distribution map exceeds a preset field intensity threshold, to obtain a potential defect area list. A three-dimensional defect positioning module is configured to acquire spatial data of the potential defect area list, construct a three-dimensional structure model inside the connector, perform structure analysis and abnormal point positioning, and obtain accurate defect position coordinates. An evolution path prediction module is configured to acquire multi-time observation values of the accurate defect position coordinates, perform time series evolution mode learning and future trend prediction on the observation values, and obtain a future evolution path of the defect. A dynamic field intensity updating module is configured to calculate the gradient change of the spatial field intensity distribution map, combine the future evolution path for comprehensive modeling, and obtain a dynamic field intensity updating sequence. A quality evaluation and reporting module is configured to perform diffusion rate calculation and quality evaluation according to the dynamic field strength update sequence and in combination with the real-time acquired field strength distribution data, and obtain a final quality detection report.

[0069] It should be noted that the FPC connector quality detection system based on multi-modal fusion provided by the embodiments of the present application is used to perform all process steps of the FPC connector quality detection method based on multi-modal fusion provided by the above embodiments, and the working principles and advantages of the two are one-to-one correspondence, thus not being described in detail.

[0070] The embodiments of the present application further provide an electronic device. The electronic device comprises a processor, a memory, and a computer program stored in the memory and capable of running on the processor, such as a dynamic field strength update program. The processor implements the steps in the above various FPC connector quality detection methods based on multi-modal fusion when executing the computer program, such as the step S11 shown in the above embodiments. Figure 1 Alternatively, the processor implements the functions of the modules / units in the above various device embodiments when executing the computer program, such as the three-dimensional defect positioning module.

[0071] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the electronic device.

[0072] The electronic device can be a desktop computer, a notebook computer, a palm computer, a smart tablet and the like. The electronic device can include, but is not limited to, a processor, a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device, and can include more or fewer components than the above, or combine certain components, or different components, for example, the electronic device can also include an input / output device, a network access device, a bus and the like.

[0073] The processor can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The processor is a control center of the electronic device, and connects various parts of the electronic device through various interfaces and lines.

[0074] The memory can be used to store the computer program and / or modules, and the processor realizes various functions of the electronic device by running or executing the computer program and / or modules stored in the memory, and calling data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required by a function (such as a sound playing function, an image playing function, etc.), etc.; and the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a nonvolatile memory, for example, a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state memory device.

[0075] The modules / units integrated in the electronic device, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can implement the steps of each method embodiment when executed by a processor. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer readable medium can include any entity or device, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. that can carry the computer program code. It should be noted that the contents included in the computer readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0076] It should be noted that the above-described device embodiments are only schematic, and the units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. In addition, the connection relationship between the modules in the device embodiment provided by the present application indicates that there is a communication connection between them, which can be realized as one or more communication buses or signal lines. Those skilled in the art can understand and implement it without creative labor.

[0077] The above-described specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the present application. It should be understood that the above-described specific embodiments are only for the specific embodiments of the present application and do not limit the protection scope of the present application. It is particularly pointed out that any modification, equivalent replacement, improvement, etc. made by those skilled in the art within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A quality inspection method for FPC connectors based on multimodal fusion, characterized in that, include: Obtain the raw multimodal dataset inside the connector, which includes electrical stress, thermal stress, and mechanical stress data; The original multimodal dataset is subjected to mean filtering and temporal alignment to obtain a standardized multiphysics sequence; Spatial feature extraction and temporal reconstruction are performed on the standardized multiphysics sequence to obtain a spatial field intensity distribution map; If the field strength distribution in the spatial field strength distribution map exceeds the preset field strength threshold, density clustering and spatial correlation analysis are performed to obtain a list of potential defect areas. Spatial data of the potential defect area list is obtained, a three-dimensional structural model of the connector is constructed, structural analysis and anomaly point localization are performed, and the precise defect location coordinates are obtained. The precise defect location coordinates are obtained from multi-time observations. The observations are then used to learn the temporal evolution pattern and predict future trends to obtain the future evolution path of the defect. The gradient change of the spatial field intensity distribution map is calculated, and a comprehensive model is performed in conjunction with the future evolution path to obtain a dynamic field intensity update sequence; Based on the dynamic field strength update sequence and combined with the real-time acquired field strength distribution data, the diffusion rate is calculated and the quality is assessed to obtain the final quality inspection report.

2. The FPC connector quality inspection method based on multimodal fusion as described in claim 1, characterized in that, The step of performing mean filtering and temporal alignment on the original multimodal dataset to obtain a standardized multiphysics sequence includes: The original multimodal dataset is smoothed using a mean filter method to obtain a denoised dataset. The signals of different modes in the denoised dataset are time-stamped and synchronized, and the missing values ​​in the synchronization process are filled in by linear interpolation to obtain a time-synchronized dataset. The time-synchronized dataset is normalized to map data with different physical dimensions to a unified scale, resulting in a standardized multiphysics sequence.

3. The FPC connector quality inspection method based on multimodal fusion as described in claim 1, characterized in that, The step of extracting spatial features and constructing a temporal sequence from the standardized multiphysics field sequence to obtain a spatial field strength distribution map includes: The standardized multiphysics sequence is processed using a convolutional neural network, and multidimensional spatial features are extracted to generate a preliminary field strength distribution map. Calculate the field strength deviation value of the preliminary field strength distribution map, and optimize the parameters of the convolutional neural network using the gradient descent method based on the field strength deviation value to obtain the optimized field strength distribution map; The optimized field strength distribution map is clustered and grouped, the spatial correlation between each group region is calculated, and the spatial correlation is reduced by principal component analysis to obtain the dimensionality-reduced feature set. The reduced feature set is stored in a pre-established distributed database and organized using a time series index to obtain a spatial field strength distribution map.

4. The FPC connector quality inspection method based on multimodal fusion as described in claim 1, characterized in that, If the field strength distribution in the spatial field strength distribution map exceeds a preset field strength threshold, density clustering and spatial correlation analysis are performed to obtain a list of potential defect areas, including: The spatial field strength distribution map is preprocessed with median filtering, and it is determined whether the preprocessed data value exceeds the preset field strength threshold. Areas exceeding the threshold are identified as abnormal areas. Density clustering is used to divide the abnormal regions, grouping regions with high field strength values ​​and spatial density to obtain a set of high-density scene regions; The feature vectors of each region in the high-density scene region set are extracted, and the spatial correlation between each region is calculated using the cosine similarity method to obtain the correlation matrix; The multidimensional features in the correlation matrix are reduced using linear discriminant analysis to obtain a list of potential defect regions.

5. The FPC connector quality inspection method based on multimodal fusion as described in claim 1, characterized in that, The process of obtaining spatial data of the potential defect area list, constructing a three-dimensional structural model of the connector's interior, performing structural analysis and anomaly point localization, and obtaining precise defect location coordinates includes: Based on the list of potential defect areas, stereoscopic microscopic 3D scanning is performed on the target area to generate 3D point cloud data; The three-dimensional point cloud data is subjected to data fusion processing to integrate multimodal data such as field strength distribution and temperature signal to construct a preliminary three-dimensional structural model of the connector. The preliminary three-dimensional structural model is subjected to structural analysis and feature extraction to identify key structural features and determine geometric anomalies. The coordinates of the geometric anomalies are calculated and corrected to obtain the precise coordinates of the defect locations.

6. The FPC connector quality inspection method based on multimodal fusion as described in claim 1, characterized in that, The process of obtaining multi-time observations of the precise defect location coordinates, performing temporal evolution pattern learning and future trend prediction on the observations to obtain the future evolution path of the defect includes: The observations at multiple times are sorted by timestamp to obtain an ordered sequence of observations. The difference between adjacent observations in the ordered sequence is calculated, and all difference values ​​exceeding a preset change threshold are collected to form a defect change increment sequence. The defect change increment sequence is input into the time series model for layer-by-layer propagation and learning to obtain a state hidden representation that characterizes the overall evolution pattern of the sequence; Based on the state hiding representation, the direction of trend continuation is determined and extrapolated to obtain the future evolution path of the defect.

7. The FPC connector quality inspection method based on multimodal fusion as described in claim 1, characterized in that, The process of calculating the gradient change of the spatial field strength distribution map and combining it with the future evolution path to perform comprehensive modeling yields a dynamic field strength update sequence, including: Gradient calculation is performed on the spatial field strength distribution map to extract the rate of change of field strength in space and obtain a gradient value sequence; Determine whether the difference between adjacent values ​​in the gradient value sequence exceeds a preset gradient stability threshold. If it exceeds the threshold, merge multiple possible branches of the future evolution path to obtain a fused path branch sequence. Based on the fusion path branch sequence, the thermal stress concentration area of ​​the non-uniform region is located, and the fluctuation sequence of real-time environmental variables is determined. The gradient value sequence, the location data of the thermal stress concentration area, and the fluctuation sequence of the environmental variables are fused to finally generate the dynamic field strength update sequence.

8. The FPC connector quality inspection method based on multimodal fusion as described in claim 1, characterized in that, The process of calculating the diffusion rate and assessing the quality based on the dynamic field strength update sequence and real-time acquired field strength distribution data, to obtain the final quality inspection report, includes: The dynamic field strength update sequence is applied to the real-time acquired field strength distribution data to generate predictive enhanced field strength distribution data. Spatial features are extracted from the predicted enhanced field strength distribution data, and the changes of the features over time are calculated to obtain the defect diffusion rate vector; Three-dimensional modeling is performed based on the defect diffusion rate vector, and environmental fluctuation and thermal stress data are fused to generate an enhanced three-dimensional coordinate sequence. The enhanced three-dimensional coordinate sequence and the defect diffusion rate vector are compared with the preset quality standard to calculate the specific values ​​of the geometric dimensions and diffusion rates that exceed the standard limits, and finally a quality inspection report is generated.

9. A quality inspection system for FPC connectors based on multimodal fusion, characterized in that, include: The multimodal data acquisition module is used to acquire the original multimodal dataset inside the connector, which includes electrical stress, thermal stress and mechanical stress data; The data preprocessing module is used to perform mean filtering and time-series alignment on the original multimodal dataset to obtain a standardized multiphysics sequence; The field strength map construction module is used to extract spatial features and construct temporal sequences from the standardized multiphysics field sequence to obtain a spatial field strength distribution map. The defect detection module is used to perform density clustering and spatial correlation analysis to obtain a list of potential defect areas if the field strength distribution in the spatial field strength distribution map exceeds a preset field strength threshold. The three-dimensional defect location module is used to acquire spatial data of the potential defect area list, construct a three-dimensional structural model of the connector's internal structure, perform structural analysis and anomaly point location, and obtain precise defect location coordinates. The evolution path prediction module is used to obtain multi-time observation values ​​of the precise defect location coordinates, perform temporal evolution pattern learning and future trend prediction on the observation values, and obtain the future evolution path of the defect. The dynamic field strength update module is used to calculate the gradient change of the spatial field strength distribution map, and to perform comprehensive modeling in conjunction with the future evolution path to obtain the dynamic field strength update sequence. The quality assessment and reporting module is used to calculate the diffusion rate and assess the quality based on the dynamic field strength update sequence and the real-time acquired field strength distribution data, and to obtain the final quality inspection report.

Images