Method for predicting service life of FPC connector

Abstract

The invention relates to an electronic connector reliability prediction technology, in particular to a method for predicting the service life of an FPC connector, and aims to solve the problem that a laboratory accelerated aging test method generally adopted in the current industry, such as a temperature and humidity cycle experiment and a mechanical plugging cycle experiment, is usually carried out under a highly controlled single stress condition, so that the service life of the FPC connector cannot be predicted. Complex multi-physical field coupling effects in a real use environment, such as dynamic interaction of random vibration, temperature shock and humidity fluctuation, cannot be reproduced. According to the method, the mechanical response signal and the electrical response signal are obtained, and the pre-constructed coupling correlation model is used for predicting the service life, so that the problem of low prediction precision caused by dependence on a single physical quantity in the prior art is solved, and the method has the advantage of improving the prediction precision and reliability.

Description

Technical Field

[0001] This application relates to electronic connector reliability prediction technology, and more specifically, to an FPC connector life prediction method. Background Technology

[0002] In the fields of consumer electronics and industrial control, FPC connectors, as core interconnect components, play a crucial role in ensuring the long-term stable operation of equipment through lifespan prediction. Current industry-standard accelerated aging test methods, such as temperature and humidity cycling and mechanical insertion / removal cycling tests, are typically conducted under highly controlled single stress conditions. These methods cannot reproduce the complex multi-physics coupling effects in real-world usage environments, such as the dynamic interactions of random vibration, sudden temperature changes, and humidity fluctuations.

[0003] The significant disconnect between these testing conditions and actual working conditions makes it difficult for test data to accurately reflect the product's lifespan performance in real-world scenarios. This puts companies in a dilemma: on the one hand, over-design may increase material and manufacturing costs, while on the other hand, underestimating the actual failure risk may lead to sudden product malfunctions during the after-sales phase, resulting in customer complaints and damage to brand reputation.

[0004] Furthermore, traditional monitoring technologies primarily rely on single physical parameters, such as contact resistance or insertion / extraction force. These parameters change extremely slowly during the early stages of connector micro-damage, making it difficult for existing sensor systems to effectively identify such weak signals. This results in a failure to provide early warnings during the initial stages of failure. Consequently, maintenance strategies have long been in a reactive state, with equipment often only being discovered after complete failure, leading to production interruptions, soaring maintenance costs, and decreased user satisfaction.

[0005] This problem is particularly acute for small and medium-sized enterprises with limited resources. Due to the lack of complex operating condition simulation tools and advanced data analysis platforms available to large enterprises, and the dual pressure of increasingly higher reliability requirements for FPC connectors in end products and tight R&D budgets, the market urgently needs a life assessment solution that does not require expensive equipment, is easy to implement, and has high predictive accuracy. Summary of the Invention

[0006] (a) Technical problems to be solved The purpose of this application is to provide an FPC connector lifetime prediction method, electronic device, and computer-readable storage medium, which has the advantages of improving prediction accuracy and reliability, can accurately evaluate the lifetime performance of FPC connectors in real complex environments, and avoids the problem of insufficient early failure warning caused by existing methods relying on a single physical quantity.

[0007] (II) Technical Solution This application provides a method for predicting the lifetime of an FPC connector, the technical solution of which is as follows: include: Acquire the mechanical and electrical response signals generated by the target FPC connector during operation; Based on mechanical and electrical response signals, a pre-built coupling correlation model is used to determine the lifetime prediction results of the target FPC connector. The coupling correlation model characterizes the correlation between the characteristic changes of the mechanical response signal and the characteristic changes of the electrical response signal.

[0008] Furthermore, this application also proposes that the mechanical response signal is a micro-strain signal; acquiring the mechanical response signal generated by the target FPC connector during operation includes: Micro-strain sensors are used to collect micro-strain signals of the FPC connector under insertion, removal, or vibration conditions by placing them at the junction of the flexible substrate and the metal terminal of the FPC connector.

[0009] Furthermore, this application also proposes that the electrical response signal is an impedance spectrum signal; obtaining the electrical response signal generated by the target FPC connector during operation includes: An AC excitation signal is applied to the contact interface through a signal electrode connected to the FPC connector contact interface, and its impedance spectrum signal is acquired.

[0010] Furthermore, this application also proposes determining the lifetime prediction results of the target FPC connector based on mechanical response signals and electrical response signals using a pre-built coupling correlation model, including: Extract the first feature value from the mechanical response signal and extract the second feature parameter from the electrical response signal; By inputting the first eigenvalue and the second eigenparameter into the coupling correlation model, the life status index of the FPC connector is obtained.

[0011] Furthermore, this application also proposes that the first characteristic value includes at least one of the following: the maximum strain value and the cumulative strain cycle value; The second characteristic parameter includes at least one of the following: charge transfer resistance and double-layer capacitance.

[0012] Furthermore, this application also proposes a method for constructing a pre-built coupling and correlation model, including: Multiple training samples were obtained. Each training sample included: a mechanical response signal sequence and an electrical response signal sequence collected during the testing of an FPC connector sample, as well as the actual life or failure state label corresponding to the FPC connector sample. Based on the training samples, a correlation equation is established between the feature values ​​extracted from the mechanical response signal sequence and the feature parameters extracted from the electrical response signal sequence to obtain a coupled correlation model.

[0013] Furthermore, this application also proposes that establishing the correlation equation between the feature values ​​extracted from the mechanical response signal sequence and the feature parameters extracted from the electrical response signal sequence includes: Linear regression is used to establish the fitting relationship between eigenvalues ​​and eigenparameters to obtain the correlation equation.

[0014] Furthermore, this application also proposes, and includes: Compare lifespan status indicators with preset warning thresholds; When the life status indicators reach or exceed the preset warning threshold, predictive maintenance prompts are generated.

[0015] Furthermore, this application also proposes an electronic device characterized by comprising: At least one processor; and The memory stores instructions that, when executed by at least one processor, cause at least one processor to perform the methods described above.

[0016] Furthermore, this application also proposes a computer-readable storage medium having a computer program stored thereon, characterized in that the program implements the above-described method when executed by a processor.

[0017] (III) Beneficial Effects Compared with the prior art, the beneficial effects of the present invention are as follows: This invention obtains mechanical and electrical response signals and uses a pre-built coupled correlation model to predict lifetime, thus solving the problem of low prediction accuracy caused by relying on a single physical quantity in the prior art. It has the advantages of improving prediction accuracy and reliability. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the logical framework structure of the FPC connector lifetime prediction method in Example 1; Figure 2 This is a schematic diagram of the logical framework structure of the FPC connector lifetime prediction method in Example 2. Detailed Implementation

[0020] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0021] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0022] In FPC connector life prediction technology, existing methods rely on accelerated aging tests in laboratories and monitoring of single physical quantities. Laboratory testing conditions differ systematically from actual usage environments, causing test data to fail to accurately map life characteristics under real-world conditions. The deviation between test data and actual lifespan stems from the failure to reproduce the coupling relationship between changes in mechanical stress and changes in the electrical state of the contact interface. Simultaneously, single-physical-quantity monitoring methods only collect independent physical quantity signals, failing to establish a correlation between changes in mechanical and electrical response signal characteristics. This allows early, weak failure signals to be masked by environmental noise, leading to predictions that deviate from actual lifespan. Consequently, companies face the problem of distorted reliability assessments, potentially leading to over-design to increase safety margins, thereby increasing material costs and manufacturing complexity, or underestimating failure risks, resulting in higher after-sales failure rates.

[0023] For example, in smartphone production lines for consumer electronics, FPC connectors are used to connect the display screen to the motherboard. During actual operation, the equipment undergoes frequent insertion and removal operations and environmental vibrations, generating mechanical response signals such as micro-strain. Simultaneously, oxidation and wear at the contact interface cause electrical response signals such as changes in impedance spectrum. Traditional monitoring methods only collect insertion and removal force data, failing to identify the synergistic changes between micro-strain accumulation and charge transfer resistance growth. This leads to the neglect of interface failure signals caused by structural fatigue in a single signal, resulting in connector contact failure without warning and unplanned equipment downtime.

[0024] If these issues are not addressed, biases in lifespan prediction will render predictive maintenance strategies unenforceable. Companies will be forced to adopt conservative designs to compensate for assessment uncertainties, increasing R&D and production costs. Furthermore, rising after-sales failure rates will impact end-product reliability and brand reputation, while a lack of ability to identify the primary failure factors hinders clear product optimization directions and prolongs technology iteration cycles.

[0025] Therefore, referring to Figure 1 This application provides a method for predicting the lifetime of an FPC connector, comprising the following steps: S100: Acquire the mechanical and electrical response signals generated by the target FPC connector during operation; S200. Based on mechanical and electrical response signals, the lifetime prediction results of the target FPC connector are determined using a pre-built coupling correlation model. Among them, the coupling correlation model characterizes the correlation between the characteristic changes of the mechanical response signal and the characteristic changes of the electrical response signal.

[0026] For ease of understanding, the following explains some key terms in this embodiment: FPC connectors, commonly known as flexible printed circuit board connectors, are used to achieve electrical connections and mechanical fixation between flexible printed circuit boards and other circuit boards or devices. FPC connectors are widely used in various electronic devices, and their reliability directly affects the overall performance and lifespan of the equipment.

[0027] Mechanical response signals refer to the physical deformation, vibration, displacement, or mechanical changes generated by external or internal mechanical stress during the operation of an FPC connector. These signals can reflect the mechanical condition of the connector components, such as fatigue, wear, or loosening.

[0028] Electrical response signals refer to the changes in electrical characteristics of an FPC connector during operation due to changes in the state of the contact interface. These signals can reflect the electrical state of the connector contact interface, such as oxidation, corrosion, wear, poor contact, or impedance changes.

[0029] A coupled correlation model is a pre-constructed mathematical or algorithmic model that characterizes the intrinsic relationship or mutual influence between the characteristic changes of mechanical response signals and the characteristic changes of electrical response signals. This model establishes a mapping relationship between the changes of two different physical quantities by learning from historical data.

[0030] Lifespan prediction results refer to information such as the remaining service life, health status indicators, or failure probability of an FPC connector, calculated or inferred through a coupled correlation model based on the mechanical and electrical response signals of the connector's current operating state. These results aim to provide a basis for predictive maintenance of equipment.

[0031] This application proposes a method for predicting the lifespan of an FPC connector. It obtains the mechanical and electrical response signals generated by the target FPC connector during operation, and uses a pre-built coupling correlation model based on these signals to determine the lifespan prediction result of the FPC connector, thereby achieving accurate prediction of the lifespan of the FPC connector.

[0032] Specifically, this method first acquires the mechanical and electrical response signals generated by the target FPC connector during operation. Mechanical response signals can be acquired in various ways. For example, an accelerometer can be installed near the FPC connector or on its supporting structure to collect vibration signals under vibration conditions; or a force sensor can be placed along the connector's insertion / removal path to monitor changes in insertion / removal forces. These signals reflect the mechanical stress conditions experienced by the connector in actual use. Electrical response signals can also be acquired in various ways. For example, a constant current can be applied between the FPC connector's contact terminals and the voltage drop measured to obtain changes in contact resistance; or, AC signals of different frequencies can be applied to the connector and its response current measured to obtain its impedance characteristics. These electrical signals reflect the conductivity and degree of degradation of the connector's contact interface.

[0033] Furthermore, after acquiring the mechanical and electrical response signals, this method uses a pre-built coupled correlation model to determine the life prediction results of the FPC connector based on these signals. The core of the coupled correlation model lies in characterizing the correlation between the characteristic changes of the mechanical response signal and the characteristic changes of the electrical response signal. For example, the model can establish a functional relationship between the accumulation of mechanical stress and the increase in contact resistance, or a mapping relationship between the change in vibration frequency and the change in charge transfer resistance. This correlation can be constructed by testing a large number of FPC connector samples and recording the changes in their electrical performance under different mechanical stresses, and then through data analysis and modeling techniques. For example, statistical regression analysis can be used to correlate certain statistics of the mechanical signal (such as root mean square value, peak value) with certain statistics of the electrical signal (such as average value, rate of change), thereby establishing a mathematical model that reflects the coordinated changes of both. Therefore, when new mechanical and electrical response signals are input into the model, the model can output the current health status or remaining life of the FPC connector based on the established internal correlation.

[0034] The following example will provide a more detailed explanation of the above technical solution: Suppose that in an industrial control device, an FPC connector is used to connect the core control module and the sensor array. This device operates long-term at "Location A" and is maintained by "User A". Traditional life prediction methods, such as relying solely on accelerated aging tests in a laboratory, often deviate from the failure modes of the device under actual operating conditions (e.g., continuous micro-vibration, ambient temperature fluctuations, and periodic plugging and unplugging maintenance), leading to inaccurate predictions. Furthermore, if only a single physical quantity is monitored, such as contact resistance, fatigue damage may have already occurred in the internal structure of the FPC connector before the contact resistance has changed significantly, thus failing to provide early warning.

[0035] The proposed lifespan prediction method can be applied to this scenario. First, during the operation of the FPC connector, its mechanical and electrical response signals are continuously acquired. Specifically, a miniature vibration sensor can be installed near the FPC connector inside the device to collect the vibration signals experienced by the connector during device operation in real time, serving as the mechanical response signal. Simultaneously, a miniature voltage probe is connected between the critical contact points of the FPC connector to monitor the voltage drop under operating current in real time, thereby indirectly reflecting changes in contact resistance, serving as the electrical response signal. These signals are continuously acquired and transmitted to the data processing unit.

[0036] Subsequently, the data processing unit inputs these real-time acquired mechanical and electrical response signals into a pre-constructed coupled correlation model. This coupled correlation model was previously established by learning from test data of similar FPC connectors under simulated actual working conditions. For example, during the model building phase, by applying different levels of vibration stress to multiple FPC connector samples and simultaneously monitoring their vibration signal characteristics and contact voltage drop characteristics, the model learned that when a certain characteristic of the vibration signal (e.g., the cumulative value of vibration energy) reaches a specific threshold, the fluctuation or average value of the contact voltage drop will show an accelerated upward trend. This correlation is encoded in the model.

[0037] Therefore, when User A's device is operating at Location A, the model receives real-time vibration and contact voltage drop signals. Based on the correlations within its internal representations, the model analyzes the synergy between the current trends in vibration and contact voltage drop. For example, if the model detects an accelerating cumulative effect of vibration and simultaneously observes a small but continuous increase in contact voltage drop, it determines that the synergistic change in these two signals indicates that the FPC connector is undergoing accelerated degradation. Based on this synergistic analysis, the model can output a more accurate lifetime prediction, such as predicting a Y% probability of failure within the next X days, or that its health status index has dropped to Z.

[0038] As demonstrated by the examples above, this method, by simultaneously acquiring mechanical and electrical response signals and utilizing a coupled correlation model to analyze the intrinsic relationship between them, can more comprehensively and sensitively capture the degradation process of FPC connectors. Changes in mechanical signals may indicate structural fatigue, while changes in electrical signals directly reflect the performance of the contact interface. Coupled analysis of both enables the identification of early, weak failure signals and provides life predictions that more closely match actual operating conditions, thus effectively addressing the shortcomings of traditional methods.

[0039] Based on the above examples, the FPC connector life prediction method proposed in this application demonstrates a significant technological contribution. In the industrial control equipment scenario of "Location A," traditional existing life prediction methods, such as those relying solely on accelerated aging tests in laboratories, often exhibit significant deviations from actual operating conditions. For instance, laboratory tests may not fully simulate the specific frequency vibrations or environmental humidity fluctuations experienced by the equipment during actual operation, leading to inconsistencies between the predicted lifespan and reality. This application, by directly acquiring mechanical and electrical response signals during actual operation and establishing a correlation between them using a coupled correlation model, enables the life prediction results to more accurately reflect the degradation process of the FPC connector under real operating conditions. This contrasts sharply with the problem of the disconnect between laboratory data and actual operating conditions in existing technologies, significantly improving the accuracy and practicality of the prediction.

[0040] Furthermore, existing life prediction methods based on a single physical quantity (e.g., monitoring only contact resistance or only vibration) have low sensitivity in identifying early, subtle failure signals. In the example above, if only contact voltage drop is monitored, the voltage drop may not reach the warning threshold even when microcracks or plastic deformation have already occurred in the metal terminals inside the FPC connector, thus missing the optimal maintenance opportunity. However, the method of this application acquires both mechanical response signals (such as vibration signals) and electrical response signals (such as contact voltage drop signals) simultaneously and uses a coupled correlation model to analyze the synergistic changes between the two. For example, the model can identify the correlation between specific frequency component changes in the vibration signal and small but persistent drifts in the contact voltage drop. This correlation may be considered noise or insignificant in a single signal, but becomes a clear indication of early failure under coupled analysis. Therefore, this method can achieve more sensitive identification of early failures in FPC connectors, thereby supporting predictive maintenance and avoiding unexpected equipment downtime. This represents a significant improvement over the passive maintenance mode of "discovering problems only after failures occur" in existing technologies.

[0041] In summary, this application innovatively combines mechanical and electrical response signals and constructs a coupled correlation model to reveal the intrinsic relationship between them. This not only solves the problem of traditional life prediction methods being out of touch with actual working conditions but also overcomes the limitation of low sensitivity in monitoring single physical quantities. This comprehensive prediction method provides a more accurate and forward-looking solution for the reliability management of FPC connectors.

[0042] In some of the solutions described above in this application, mechanical response signals are proposed to monitor the mechanical state of FPC connectors. However, in this process, the signal type and acquisition method are not specified, which may lead to inaccurate signal acquisition or failure to reflect the actual failure mechanism. In particular, in the critical area where the flexible substrate and metal terminal meet, and under real working conditions such as insertion, removal or vibration, the signal cannot effectively capture early weak failure characteristics, thereby affecting the reliability and accuracy of life prediction.

[0043] In response, this application further proposes a method for obtaining the mechanical response signal generated by the target FPC connector during operation, wherein the mechanical response signal is a micro-strain signal; the method for obtaining the mechanical response signal generated by the target FPC connector during operation includes: acquiring the micro-strain signal of the FPC connector under insertion / removal or vibration conditions by using a micro-strain sensor disposed at the junction of the flexible substrate and the metal terminal of the FPC connector.

[0044] Specifically, the mechanical response signal is defined as a micro-strain signal, which is a quantitative representation of the minute deformation (strain) generated inside an object when it is subjected to external force. It is typically measured in micrometers per meter (µε) and can sensitively capture subtle structural changes within materials, making it particularly suitable for monitoring early failure signs such as fatigue and damage. This micro-strain signal can be measured in various ways. For example, it can be achieved using resistance strain gauge technology; when the strain gauge deforms with the object, its resistance changes, and the micro-strain value can be calculated by measuring the change in resistance. Alternatively, it can be achieved using a fiber Bragg grating (FBG) sensor; when the fiber is strained, the grating period changes, causing a wavelength shift in the reflected light, and the micro-strain information can be obtained by detecting the amount of wavelength shift. Another option is to use a piezoelectric thin-film sensor; when the piezoelectric thin film is strained, it generates an electric charge, and the micro-strain can be reflected by measuring the amount of charge or voltage change.

[0045] The junction between the flexible substrate and the metal terminals of an FPC connector refers to the physical connection area between the flexible circuit board (usually made of flexible materials such as polyimide) and the metal terminals (such as gold-plated copper alloy terminals) used for electrical connection in the FPC connector. This area is a critical part of the FPC connector most susceptible to stress concentration, fatigue damage, and failure under mechanical stress, such as insertion / removal, bending, and vibration. Monitoring the micro-strain of this junction can directly reflect the mechanical stress state and damage accumulation of the critical structural parts of the FPC connector in actual use, providing a direct basis for assessing the mechanical reliability and predicting the lifespan of the connector. This junction can refer to the physical connection interface formed by welding, riveting, or conductive adhesive bonding between the flexible substrate and the metal terminals; or it can refer to the area on the flexible substrate that supports the metal terminals, including the transition area between the terminal root and the flexible substrate.

[0046] The micro-strain sensor is a device that converts minute mechanical strain into measurable electrical signals. It features high sensitivity, small size, and fast response, enabling precise capture of minute deformations on material surfaces. Its function is to monitor strain changes in key components of FPC connectors in real-time or near real-time, converting these physical changes into electrical signals for data acquisition and analysis. This sensor can employ resistance strain gauges, typically in the millimeter range, fixed to the measured surface with adhesive; it can also use MEMS (Micro-Electro-Mechanical Systems) strain sensors, which are even smaller and more integrated, and can be directly integrated onto or near flexible substrates; or it can utilize fiber optic strain sensors, such as fiber Bragg grating sensors, which offer advantages such as strong electromagnetic interference resistance and distributed measurement capabilities.

[0047] The mating / removal and vibration conditions refer to two typical mechanical stress environments that FPC connectors may experience in practical applications. The mating / removal condition refers to the repeated mating and removing operations during installation, maintenance, or use, causing cyclic mechanical stress on the connector contact interface and structural components. The vibration condition refers to the mechanical vibration generated by external or internal vibration sources during equipment operation, causing high-frequency, low-amplitude alternating stress on the connector structural components. Collecting signals under these conditions can simulate the stress conditions of FPC connectors in real-world usage environments, ensuring that the collected data is highly correlated with actual failure modes, thereby improving the accuracy of lifespan prediction. The mating / removal condition can be simulated using an automated mating / removal testing machine, setting specific mating / removal speeds, strokes, and number of cycles; the vibration condition can be simulated using a vibration table, setting specific vibration frequencies, amplitudes, and durations, such as sinusoidal vibration or random vibration; alternatively, signal acquisition can be triggered by monitoring the equipment's own mating / removal operations or vibration states in the actual product operating environment.

[0048] The purpose of acquiring micro-strain signals from FPC connectors under mating / removal or vibration conditions is to obtain dynamic mechanical response data of FPC connectors under actual or simulated mechanical stress. By acquiring micro-strain signals under specific conditions, information such as strain accumulation and strain amplitude changes in key parts of the connector can be captured during the stress process. This information is a key indicator for evaluating connector mechanical fatigue and predicting lifespan. This acquisition process can be achieved by connecting micro-strain sensors to a data acquisition card (DAQ). During mating / removal or vibration testing of the FPC connector, the electrical signals output by the sensors are continuously recorded at a preset sampling frequency and duration, and converted into micro-strain data. Alternatively, a wireless sensor network can be used to transmit the data from the micro-strain sensors to the receiving end via a wireless module, enabling remote or distributed data acquisition, which is particularly suitable for space-constrained or mobile devices. Furthermore, a microcontroller integrated on the FPC connector can perform preliminary processing and storage of the micro-strain sensor data, and upload the data upon specific events (such as reaching a mating / removal threshold or detecting abnormal vibration).

[0049] This application's solution explicitly defines the mechanical response signal as a micro-strain signal and, combined with targeted sensor placement and a dynamic operating condition acquisition strategy, constructs a highly accurate mechanical condition monitoring system closely related to the failure mechanism of FPC connectors. Specifically, the micro-strain signal, as a direct quantification of minute deformations within the material, can serve as a "source characterization signal" for the mechanical fatigue failure of FPC connectors, directly reflecting the damage accumulation at the junction of the flexible substrate and metal terminals under stress. This reveals the essence of failure more effectively than macroscopic mechanical signals. By precisely placing the micro-strain sensor at the junction of the flexible substrate and metal terminals of the FPC connector, this solution identifies high-incidence areas of stress concentration and fatigue failure, ensuring the locality and criticality of the acquired signals and avoiding noise or bias that may be introduced by signals from non-critical areas. Furthermore, acquiring micro-strain signals under insertion / removal or vibration conditions captures the "cyclic accumulation and gradual evolution" characteristics exhibited by these dynamic micro-strain signals. These dynamic characteristics highly match the failure process of "stress cycling leading to fatigue aging" in actual use, thus ensuring the authenticity and relevance of the acquired signals. This triple constraint of "precise signal type limitation + targeted acquisition location design + dynamic adaptation to operating conditions" enables the acquired mechanical response signals to accurately and sensitively reflect the early mechanical damage evolution of FPC connectors. This provides high-quality, highly correlated mechanical input data for subsequent determination of FPC connector life prediction results using pre-built coupled correlation models, greatly improving the reliability and accuracy of the entire life prediction method.

[0050] As one specific implementation, lifespan prediction can be performed on the FPC connector connecting the motherboard and display module in a smartphone. In this FPC connector, a miniature resistance strain gauge (e.g., 1mm x 0.5mm in size, with a resistance of 120Ω and a strain coefficient of 2.0) is precisely bonded to the junction of the flexible substrate and the metal terminal of the FPC connector, specifically the root region on the polyimide flexible substrate supporting the gold-plated copper terminal. During lifespan prediction, the smartphone can be placed on an automated test bench to simulate insertion and removal conditions. For example, a robotic arm can repeatedly insert and remove the FPC connector of the display module at a speed of 50mm / s for 10,000 cycles. During this process, the micro-strain sensor, connected to a high-precision data acquisition system (e.g., a data logger containing a multi-channel strain amplifier and a 24-bit analog-to-digital converter), continuously acquires the voltage signal output by the sensor at a sampling rate of 1kHz and converts it into micro-strain values. These timestamped micro-strain data are stored for subsequent feature extraction and lifespan prediction.

[0051] Through the above technical solution, this application effectively solves the problem of unclear signal type and acquisition method in traditional methods, significantly improving the accuracy of signal acquisition. By defining the mechanical response signal as a micro-strain signal and precisely placing its sensor at the junction of the flexible substrate and metal terminal of the FPC connector, this solution can directly capture the micro-damage evolution of the failure source of the FPC connector, thus establishing a high correlation between the acquired signal and the actual failure mechanism. In addition, acquiring micro-strain signals under real working conditions such as insertion / removal or vibration can effectively capture early and weak dynamic failure characteristics, breaking through the bottleneck of traditional methods in early failure signal identification. These improvements work together to provide high-quality and high-reliability mechanical response data for subsequent coupled correlation models, thereby greatly improving the reliability and accuracy of FPC connector life prediction, making predictive maintenance possible, and meeting the needs of small and medium-sized enterprises for low-cost, high-precision life prediction solutions.

[0052] In some of the solutions described above in this application, electrical response signals are proposed to obtain electrical status information of FPC connectors. However, in this process, the definition of electrical response signals is not specific enough, which may lead to insufficient sensitivity in identifying early weak failure signals, affecting the accuracy of life prediction and the feasibility of predictive maintenance.

[0053] In this regard, this application further proposes that the electrical response signal is an impedance spectrum signal; obtaining the electrical response signal generated by the target FPC connector during operation includes: applying an AC excitation signal to the contact interface through a signal electrode connected to the contact interface of the FPC connector, and acquiring its impedance spectrum signal.

[0054] Specifically, the electrical response signal is defined as an impedance spectroscopy signal. An impedance spectroscopy signal is a signal that characterizes the electrochemical properties of a material or interface by applying an AC excitation signal over a wide frequency range and measuring the response current. It provides detailed information about the frequency-dependent changes in parameters such as resistance, capacitance, and inductance, revealing physicochemical changes within the contact interface, such as oxide layer formation, contact area reduction, and charge transfer processes—microscopic degradation processes. This signal is highly sensitive to minute interface degradation, capable of capturing early interface changes at the atomic and micrometer levels, quantifying the physical state evolution of the contact interface, including oxide layer thickness, contact area changes, and surface wear. Besides impedance spectroscopy signals, the electrical response signal can also be an electrochemical noise signal, reflecting corrosion or failure processes by monitoring spontaneously generated current or voltage fluctuations at the contact interface.

[0055] The signal electrode is a conductive component used to apply excitation signals and / or acquire response signals. It is designed to connect directly to the male and female terminal contact interfaces of the FPC connector. This connection method ensures targeted signal acquisition, directly targeting the core failure area—the male and female terminal contact interface of the FPC connector—effectively eliminating interference from non-critical areas such as external circuits or flexible substrate wiring. The signal electrode can be implemented using redundant contacts or test points reserved in the FPC connector design, or it can be temporarily or permanently connected to the contact interface using microprobes or conductive adhesives without affecting the normal function of the FPC connector.

[0056] The AC excitation signal is an electrical signal that varies periodically over time, typically a sine wave with a specific frequency and amplitude. This AC excitation signal can penetrate insulating or semi-insulating layers such as oxide films and oil stains that may exist at the contact interface, thereby accurately acquiring the impedance response of the interface itself, rather than the equivalent impedance of the overall circuit. By changing the frequency of the excitation signal, interfacial processes at different time scales can be detected. The AC excitation signal can be generated by an electrochemical workstation or impedance analyzer, with a frequency range covering from millihertz to megahertz. Alternatively, a dedicated signal generator combined with a current / voltage amplifier can be used to generate the desired AC excitation signal.

[0057] Acquiring the impedance spectrum signal refers to measuring the voltage and current responses generated at the contact interface under AC excitation signals of different frequencies, and calculating the corresponding complex impedance values. By obtaining the electrical characteristics of the FPC connector contact interface at different frequencies, these characteristics are closely related to the physical and chemical states of the interface (such as oxide layer thickness, contact area, charge transfer rate, etc.), thus providing a data basis for subsequent lifetime prediction. Impedance spectrum signals can be acquired using an electrochemical workstation or impedance analyzer. These devices typically integrate signal generation, response measurement, and data processing functions. Alternatively, lock-in amplifier technology can be used to accurately measure impedance by simultaneously detecting the amplitude and phase difference of the excitation and response signals.

[0058] This application aims to address the insufficient sensitivity of traditional electrical signals in identifying early, weak failure signals of FPC connectors. First, the electrical response signal is explicitly defined as an impedance spectrum signal. Impedance spectrum signals provide richer interface electrical information than traditional contact resistance signals because they reveal internal physicochemical changes at the contact interface, such as oxide layer formation, contact area reduction, and charge transfer processes, by applying an AC excitation signal over a wide frequency range and measuring the response. These microscopic changes are the root cause of early failures in FPC connectors. Second, to ensure that the acquired impedance spectrum signal accurately reflects the state of the FPC connector contact interface, this application proposes acquiring the signal through a signal electrode connected to the FPC connector contact interface. This targeted connection method allows the excitation and response signals to act directly on the core failure area—the male and female terminal contact interface—effectively eliminating interference from non-critical areas such as external circuits and flexible substrate wiring, ensuring signal purity and specificity. Next, an AC excitation signal is applied to this contact interface. The characteristics of the AC excitation signal allow it to penetrate any oxide films, oil stains, or other insulating or semi-insulating layers present at the contact interface, thereby detecting the true electrical response of the interface itself.

[0059] By dynamically applying and acquiring data during operation, the interface state changes of the FPC connector under actual working conditions can be monitored in real time. Ultimately, the acquired impedance spectrum signal contains rich electrical degradation information of the FPC connector contact interface. This information is highly sensitive to even minor interface degradation and can detect early signs of failure. These high-quality electrical response signals, along with the mechanical response signals obtained using the aforementioned method, serve as input to a pre-constructed coupled correlation model. The coupled correlation model can characterize the correlation between the characteristic changes of the mechanical response signal and the characteristic changes of the electrical response signal, thereby more accurately and comprehensively assessing the lifespan of the FPC connector and achieving high-precision lifespan prediction. In this way, this scheme not only solves the problems of ambiguous electrical signal definitions and insufficient sensitivity, but more importantly, it provides high-quality, high-sensitivity electrical input to the coupled correlation model, enabling the model to more effectively capture early failure signs of the FPC connector during operation, thus significantly improving the accuracy of the overall lifespan prediction method and the feasibility of predictive maintenance.

[0060] As a specific implementation, the electrical response signal can specifically refer to the complex impedance data obtained through electrochemical impedance spectroscopy (EIS) at the FPC connector contact interface. This data is typically presented in the form of a Nyquist plot or a Bode plot, containing key parameters such as charge transfer resistance and double-layer capacitance. To connect the signal electrodes to the FPC connector contact interface, additional test contacts not used for normal signal transmission can be reserved near the contact areas of the male and female terminals during the FPC connector design phase. These test contacts can serve as signal electrodes, connected to external testing equipment via microwires or flexible circuit boards. Alternatively, for existing FPC connectors, a high-precision probe station can be used to precisely contact the contact surfaces of the male and female terminals of the FPC connector as signal electrodes. When applying an AC excitation signal to the contact interface, a portable electrochemical workstation can be used to apply a sinusoidal AC voltage signal with an amplitude of 10mV and a frequency range from 100kHz to 10mHz to the FPC connector contact interface through the aforementioned signal electrodes. While applying an AC excitation signal, the electrochemical workstation simultaneously measures the AC current response flowing through the contact interface and calculates the complex impedance value corresponding to each frequency point according to the Ω law. These complex impedance values ​​constitute an impedance spectrum signal and are recorded for subsequent analysis.

[0061] By concretizing electrical response signals into impedance spectroscopy signals and employing a targeted acquisition method, this approach significantly improves the sensitivity of identifying early, weak failure signals at the FPC connector contact interface. Impedance spectroscopy signals can capture microscopic interface changes that are difficult to detect with traditional electrical parameters, such as the initial formation of oxide layers and minor wear at contact points, thus overcoming the bottleneck in early failure signal identification. By connecting the signal electrodes to the FPC connector contact interface and applying an AC excitation signal, the authenticity and relevance of the acquired signals are ensured, effectively eliminating external circuit interference and allowing the data to directly reflect the electrical state of the core failure area—the contact interface. This provides high-quality, high-sensitivity electrical input for the pre-built coupling correlation model in the aforementioned method, enabling the model to more accurately characterize the correlation between mechanical and electrical responses. Therefore, this approach can capture early signs of FPC connector failure, providing a sufficient time window for predictive maintenance, thereby significantly improving the accuracy of life prediction and the feasibility of predictive maintenance, and effectively avoiding equipment downtime losses caused by underestimating failure risks.

[0062] In some of the solutions described above in this application, a pre-built coupled correlation model is proposed to determine the lifetime prediction results in order to improve the lifetime prediction accuracy. However, in its implementation, directly using the original mechanical response signal and electrical response signal may lead to low model processing efficiency or failure to effectively capture key features, thereby affecting the accuracy of prediction and the sensitivity of early failure identification.

[0063] In response, this application further proposes to determine the lifetime prediction result of the target FPC connector based on the mechanical response signal and the electrical response signal using a pre-constructed coupling correlation model, including: extracting a first feature value from the mechanical response signal and extracting a second feature parameter from the electrical response signal; inputting the first feature value and the second feature parameter into the coupling correlation model to obtain the lifetime status index of the FPC connector.

[0064] Specifically, extracting the first feature value from the mechanical response signal aims to identify and quantify key information directly related to the mechanical degradation and fatigue damage of the FPC connector from the raw mechanical response signal generated during operation. This process effectively filters out noise and redundant data in the signal, focusing on mechanical change indicators that have a decisive impact on lifespan prediction. For example, time-domain analysis methods can be used to calculate statistical characteristics such as peak value, root mean square value, variance, or kurtosis of the signal; or frequency-domain analysis techniques, such as Fourier transform or wavelet analysis, can be used to extract features such as energy distribution, dominant frequency components, or harmonic content of the signal within a specific frequency range. These extracted feature values ​​can more concisely and accurately characterize the mechanical state of the FPC connector. Extracting the second feature parameter from the electrical response signal aims to identify and quantify key information related to the degradation, corrosion, or wear of the electrical performance of the connector contact interface from the raw electrical response signal generated during operation.

[0065] This process allows for the extraction of core parameters reflecting the electrical health of FPC connectors from complex electrical signals. For example, equivalent circuit models can be fitted to impedance spectrum signals to extract parameters of circuit elements with clear physical meanings, such as charge transfer resistance and double-layer capacitance. Alternatively, the impedance magnitude and phase angle changes of the impedance spectrum curve at specific frequency points or ranges can be analyzed; these changes are often closely related to the microstructure and chemical state of the contact interface. These extracted feature parameters can more effectively reflect the degree of electrical degradation of the FPC connector. Inputting the first feature value and the second feature parameter into the coupling correlation model involves using the highly representative and low-dimensional mechanical and electrical feature data obtained through the above feature extraction process as input to the pre-constructed coupling correlation model.

[0066] This approach avoids the computational complexity and data redundancy issues associated with directly processing raw high-dimensional signals, enabling the model to learn and utilize the intrinsic correlation between mechanical and electrical features more efficiently and accurately. Specifically, the extracted first and second feature parameters can be integrated into a feature vector or matrix, which is then passed to the model for processing through its input interface. The resulting lifespan status index of the FPC connector is obtained by the coupled correlation model based on the input mechanical first feature value and electrical second feature parameter. After internal correlation analysis and calculation, it outputs a quantified numerical value or status indicator to assess the current health status or remaining service life of the FPC connector. This lifespan status index is not a simple reflection of a single feature, but rather comprehensively considers the synergistic effect of both mechanical and electrical degradation mechanisms, providing a more comprehensive and accurate reflection of the overall lifespan status of the FPC connector. For example, the index can be a continuous value between 0 and 1, where 1 represents brand new and 0 represents complete failure; or it can be a discrete level classification, such as "healthy," "mildly aged," "moderately aged," or "severely aged."

[0067] The solution proposed in this application addresses the inefficiencies and insufficient feature capture that may occur when directly using raw signals for life prediction by refining the mechanical and electrical response signals generated by the FPC connector during operation. Specifically, firstly, first feature values ​​are extracted from the raw mechanical response signals. These feature values ​​are selected and quantified, representing key information that can effectively characterize the mechanical damage and fatigue of the FPC connector.

[0068] Simultaneously, second feature parameters are extracted from the original electrical response signal. These parameters are refined and summarized core indicators that accurately reflect the degree of electrical performance degradation at the FPC connector contact interface. This two-dimensional feature extraction transforms the high-dimensional, complex original signal into low-dimensional, refined feature data, significantly reducing the complexity of subsequent model processing and enhancing the correlation between the data and the FPC connector failure mechanism. Subsequently, these extracted first and second feature values ​​are used as inputs to a pre-constructed coupled correlation model. This coupled correlation model characterizes the correlation between the feature changes of the mechanical response signal and the feature changes of the electrical response signal. Based on these refined input features, it can deeply explore the synergistic effect and mutual influence between mechanical and electrical degradation. Finally, through the analysis of these coupled features, the model outputs a comprehensive FPC connector life status index. This index not only quantifies the current health status of the FPC connector but also more accurately predicts its remaining life, thus providing a reliable basis for predictive maintenance.

[0069] This solution provides high-quality, highly correlated input to the coupling correlation model through feature extraction, enabling the model to more effectively capture the multi-physics coupling failure law of FPC connectors under complex working conditions, and significantly improve the accuracy of lifetime prediction and the sensitivity of early failure identification.

[0070] The following is a concrete example. Suppose that during the operation of the FPC connector, micro-strain signals are acquired using micro-strain sensors, and impedance spectrum signals are acquired using signal electrodes. To extract key information from these raw signals, the following steps can be taken: For the micro-strain signals, the maximum strain value within each insertion / removal or vibration cycle can be calculated and accumulated to obtain the cumulative strain cycle value, which serves as the first feature value. These values ​​directly reflect the mechanical load and cumulative damage experienced by the FPC connector. For the impedance spectrum signal, an equivalent circuit model (e.g., the R(Q(R(L(R)))) model) can be used to fit it, thereby extracting charge transfer resistance and double-layer capacitance as second feature parameters. These parameters effectively characterize the corrosion degree and electrochemical activity of the FPC connector contact interface. Subsequently, these extracted maximum strain values, cumulative strain cycle values, charge transfer resistance, and double-layer capacitance are combined into a feature vector and input into a pre-trained coupling correlation model. This coupling correlation model can be a model based on a support vector machine (SVM) or a neural network (NN), which has learned the complex relationship between these features and the actual lifespan of the FPC connector. After receiving these feature vectors, the model performs internal calculations and ultimately outputs a value between 0 and 100. For example, this value represents the percentage of the FPC connector's remaining lifespan. An output of 85 indicates that the FPC connector still has 85% of its lifespan remaining, while an output of 20 indicates that the FPC connector is nearing the end of its lifespan and requires warning or maintenance.

[0071] Through the above technical solution, this application effectively solves the problem that directly using the original mechanical and electrical response signals for lifetime prediction may lead to low model processing efficiency or an inability to effectively capture key features, thereby affecting prediction accuracy and sensitivity to early failure identification. Specifically, by extracting a first feature value from the mechanical response signal and a second feature parameter from the electrical response signal, this solution can transform high-dimensional, complex raw data into low-dimensional, refined core features that are highly correlated with the failure mechanism of FPC connectors. This not only significantly reduces the computational load of the coupled correlation model, making it easier to deploy and run with the limited computing resources of small and medium-sized enterprises, but also greatly improves the model's ability to identify early, weak failure signals such as mechanical fatigue and electrical degradation of FPC connectors.

[0072] Based on these high-quality input features, the pre-built coupled correlation model can more accurately capture the synergistic effect between mechanical and electrical degradation, thus outputting more accurate and reliable life status indicators. Furthermore, since the life status indicators are based on eigenvalues ​​and characteristic parameters with clear physical meaning, the prediction results have better interpretability, helping engineers analyze the failure modes of FPC connectors and providing strong support for product design optimization and fault tracing. Overall, this solution optimizes data input, enabling the coupled correlation model to work more efficiently and accurately, thereby achieving accurate prediction and early warning of FPC connector life, effectively avoiding over-design or equipment downtime losses caused by inaccurate predictions of traditional methods.

[0073] In some of the solutions described above in this application, a first feature value and a second feature parameter are extracted and used as inputs to a coupled correlation model to predict lifetime. However, if the selection of the feature value is not specific, it may fail to effectively capture key failure signals, resulting in insufficient prediction sensitivity and affecting the accuracy of lifetime prediction and the ability to identify early failures. To address this, this application further proposes that the first feature value includes at least one of the following: maximum strain value, cumulative strain cycle value; and the second feature parameter includes at least one of the following: charge transfer resistance, electric double layer capacitance.

[0074] The maximum strain value refers to the instantaneous maximum deformation experienced by the interface between the flexible substrate and the metal terminals of the FPC connector during operation. This value directly reflects the peak stress intensity experienced by the connector under extreme conditions such as insertion / removal, vibration, or impact. It can be obtained by real-time monitoring of strain signals using micro-strain sensors and extracting the peak value from the signal waveform; or by simulating the stress distribution under specific conditions through finite element analysis and determining the peak value of the maximum strain region. The cumulative strain cycle value refers to the cumulative effect of the number of strain cycles and their amplitude experienced by the interface between the flexible substrate and the metal terminals of the FPC connector during long-term operation. This value is a key indicator for assessing material fatigue damage and progressive degradation. It can be obtained by continuously acquiring strain signals using micro-strain sensors and combining them with fatigue analysis algorithms such as rainflow counting to count and accumulate strain cycles; or by recording the number of insertion / removal cycles, vibration duration, and other operating data of the connector, and estimating it using a preset strain-life curve.

[0075] Charge transfer resistance (CRT) refers to the resistance encountered when charge transfers from one electrode to another at the contact interface of an FPC connector. This resistance value is closely related to the electrochemical state of the contact interface, including oxide layer thickness, corrosion degree, and interfacial activity, and is a sensitive indicator of early oxidation and electrochemical degradation of the contact interface. It can be obtained through electrochemical impedance spectroscopy (EIS), which involves applying AC excitation signals of different frequencies to the contact interface, measuring its impedance response, and then fitting the impedance spectrum data using an equivalent circuit model to extract the CRT value; or through electrochemical methods such as cyclic voltammetry, analyzing electrode reaction kinetics to indirectly deduce the CRT. Double-layer capacitance refers to the capacitance formed at the contact interface of an FPC connector due to the redistribution of charge between the conductor surface and the electrolyte solution.

[0076] This capacitance value is closely related to the physical states of the interface, such as the effective contact area, surface roughness, and microstructure. It is an effective indicator for detecting fretting wear, changes in contact pressure, and interface morphology degradation. It can be obtained through electrochemical impedance spectroscopy (EIS), by fitting the impedance spectrum data with an equivalent circuit model to extract the double-layer capacitance value; or through transient electrochemical techniques such as the constant potential step method, measuring the current response curve, and then calculating the double-layer capacitance.

[0077] This application aims to provide more targeted and relevant inputs for FPC connector life prediction by specifically defining the types of the first characteristic value and the second characteristic parameter. During operation, FPC connectors are subjected to the combined effects of mechanical stress (such as mating and vibration) and environmental factors (such as oxidation and wear), leading to gradual performance degradation. To comprehensively capture these degradation processes, this solution selects the maximum strain value and the cumulative strain cycle value as key features of the mechanical response signal, and charge transfer resistance and double-layer capacitance as key features of the electrical response signal. Specifically, the maximum strain value can reflect the instantaneous maximum deformation experienced by the connector under extreme operating conditions, which is crucial for identifying sudden mechanical damage. The cumulative strain cycle value focuses on quantifying the fatigue accumulation effect caused by stress cycles during long-term use, thereby comprehensively assessing progressive mechanical degradation. These two mechanical features complement each other, jointly covering the full-dimensional mechanical failure modes of FPC connectors from sudden damage to progressive degradation.

[0078] Meanwhile, charge transfer resistance is highly sensitive to minute changes in the oxide layer thickness at the contact interface, effectively indicating early oxidation risks and playing a crucial role in capturing precursors of failure such as electrochemical corrosion and environmental oxidation. Double-layer capacitance reflects changes in the effective contact area, roughness, and contact pressure at the contact interface, offering significant advantages in detecting physical interface degradation such as fretting wear and poor contact. The synergistic effect of these two electrical features allows for precise quantification of the microscopic degradation process at the contact interface, overcoming the bottleneck of traditional single electrical features in early failure identification. By inputting these carefully selected mechanical and electrical features with clear physical meaning and failure correlation into a pre-constructed coupled correlation model, the model can more accurately uncover the intrinsic causal relationship between mechanical and electrical degradation. For example, an increase in the cumulative strain cycle value may lead to fluctuations in terminal contact pressure, thereby affecting double-layer capacitance and accelerating interface oxidation, resulting in an increase in charge transfer resistance. This synergistic analysis of multi-dimensional features enables the coupled correlation model to establish a more accurate lifetime prediction model, effectively solving the problems of fuzzy feature selection and insufficient prediction sensitivity in traditional methods, and providing highly correlated and complementary input features for FPC connector lifetime prediction.

[0079] As a specific implementation method, the characteristic values ​​and parameters can be acquired and extracted during the operation of the FPC connector in the following ways. For mechanical response signals, micro-strain sensors installed at the junction of the flexible substrate and the metal terminals of the FPC connector can be used to continuously collect micro-strain signals of the connector under insertion / removal or vibration conditions. In the acquired micro-strain signal sequence, signal processing algorithms can be used to identify and extract the peak strain in each strain cycle, which is taken as the maximum strain value. Simultaneously, rainflow counting can be used to process the entire strain signal sequence, count the number of cycles at different strain amplitudes, and accumulate the cumulative strain cycle value. For electrical response signals, a series of AC excitation signals of different frequencies can be applied to the contact interface through signal electrodes connected to the FPC connector contact interface, and the corresponding impedance response can be acquired simultaneously to obtain impedance spectrum data.

[0080] Subsequently, using specialized electrochemical impedance spectroscopy (EIS) analysis software, the acquired impedance spectral data was fitted to a pre-defined equivalent circuit model (e.g., the Randles circuit model or a circuit model containing constant phase angle elements). This fitting process allows for the direct extraction of specific values ​​for charge transfer resistance and double-layer capacitance from the parameters of the equivalent circuit model. For example, in the Randles circuit model, charge transfer resistance typically corresponds to a resistive element connected in parallel with the double-layer capacitance, while the double-layer capacitance directly corresponds to this parallel capacitive element. In this way, key mechanical and electrical characteristics for lifetime prediction can be efficiently and accurately obtained from the actual operating data of the FPC connector.

[0081] Through the above technical solution, this application significantly improves the accuracy of FPC connector life prediction and early failure identification capability by precisely defining the types of the first characteristic value and the second characteristic parameter. Specifically, the combination of the maximum strain value and the cumulative strain cycle value enables this solution to comprehensively cover all dimensions of mechanical failure modes of FPC connectors, from sudden mechanical damage (such as insertion and extraction impact) to progressive fatigue degradation (such as long-term vibration), avoiding the one-sidedness of a single mechanical feature and ensuring the comprehensiveness of mechanical damage data. At the same time, the synergistic application of charge transfer resistance and double-layer capacitance enables this solution to accurately quantify the micro-degradation process of the contact interface. The charge transfer resistance is highly sensitive to small changes in the early oxide layer thickness, and can capture early signs of failure in the initial stage of oxidation; the double-layer capacitance can effectively identify abnormal physical states of the interface caused by fretting wear, contact pressure fluctuations, etc.

[0082] This dual-dimensional, complementary feature selection not only significantly improves the sensitivity to identify early, weak failure signals, but also provides highly correlated and complementary input features for the coupled correlation model. This allows the model to more accurately uncover the intrinsic causal relationship between mechanical and electrical failures, thereby achieving precise prediction of FPC connector lifespan. Furthermore, these defined features are easily extracted from the original signal without complex algorithms, which aligns perfectly with lightweight modeling methods such as linear regression. This effectively lowers the technical implementation threshold and cost, perfectly meeting the needs of small and medium-sized enterprises for low-cost, high-precision lifespan prediction solutions, and solving the problems of insufficient prediction sensitivity and high R&D costs in traditional methods.

[0083] In some of the solutions described above in this application, a pre-built coupled correlation model is proposed to combine mechanical response signals and electrical response signals for life prediction. However, in this process, the specific method of model construction is not clear, which may lead to a lack of real working condition data support for model training, and an inability to effectively capture the intrinsic correlation between changes in mechanical characteristics and changes in electrical characteristics, thereby affecting the accuracy and reliability of life prediction.

[0084] In response, this application further proposes a method for constructing a pre-built coupling correlation model, which includes: acquiring multiple sets of training samples, each set of training samples including: a mechanical response signal sequence and an electrical response signal sequence collected during the testing of an FPC connector sample, and the actual life or failure state label corresponding to the FPC connector sample; based on the training samples, establishing a correlation equation between the feature values ​​extracted from the mechanical response signal sequence and the feature parameters extracted from the electrical response signal sequence to obtain the coupling correlation model.

[0085] The acquisition of multiple training samples aims to provide a sufficient data foundation for constructing the coupled correlation model, enabling it to learn the failure modes and life evolution patterns of FPC connectors under different operating conditions. This can be achieved by conducting accelerated aging tests on multiple FPC connector samples in a laboratory environment, simulating different insertion / removal, vibration, and temperature cycles, and continuously collecting data during this process; or by conducting long-term monitoring of a batch of FPC connectors in real-world application scenarios, collecting data on their performance under actual operating conditions, and recording their failure time or state. Each training sample includes the mechanical and electrical response signal sequences collected during the testing of one FPC connector sample, as well as the actual life or failure state label corresponding to the FPC connector sample. This sample structure is one of the core elements of this scheme. The mechanical and electrical response signal sequences provide dynamic changes in the FPC connector throughout the entire testing or operating cycle, rather than static data at a single point in time, which is crucial for capturing the dynamic process of failure evolution. The actual life or failure state label provides the model with a supervised learning target, enabling it to establish a direct mapping relationship between changes in signal characteristics and the life consumption or final failure state of the FPC connector. For example, the mechanical response signal sequence can be a curve showing the change of strain values ​​over time, continuously acquired by a micro-strain sensor; the electrical response signal sequence can be a curve showing the change of impedance values ​​over time, obtained by scanning at different frequencies using an impedance spectrometer. The actual lifespan label can be the number of mating / removal cycles or operating time when the FPC connector reaches a preset failure standard; the failure state label can be discrete states such as "normal," "early degradation," or "near failure." Based on training samples, a correlation equation is established between the feature values ​​extracted from the mechanical response signal sequence and the feature parameters extracted from the electrical response signal sequence. This step aims to quantify the mutual influence and synergistic changes of mechanical and electrical responses during the lifespan evolution of the FPC connector by analyzing the training sample data. By establishing the correlation equation, the model can understand how electrical performance changes accordingly when mechanical performance degrades, and vice versa, thus forming a coupled failure prediction mechanism. For example, multiple regression analysis can be used, with multiple feature values ​​(such as strain amplitude and cumulative strain) extracted from the mechanical response signal sequence as independent variables and feature parameters (such as charge transfer resistance and double-layer capacitance) extracted from the electrical response signal sequence as dependent variables, to establish a multi-input single-output or multi-output mathematical model. Alternatively, time series analysis methods, such as dynamic time warping (DTW) or hidden Markov models (HMM), can be used to capture the temporal correlation between mechanical and electrical feature sequences and construct equations or models that can describe this dynamic correlation. Ultimately, a coupled correlation model is obtained, which is the core output of this scheme. It is a mathematical model or algorithm that can characterize the correlation between changes in mechanical response signal features and changes in electrical response signal features.In actual prediction, this model can comprehensively consider the mechanical and electrical conditions of FPC connectors, thus providing more comprehensive and accurate lifespan prediction results. For example, the model can be a prediction model built based on statistical principles. Through the aforementioned correlation equation, combined with the actual lifespan or failure state labels of FPC connector samples, a regression model or classification model that can output lifespan status indicators can be further trained. Alternatively, the model can also be a model built based on machine learning algorithms, such as Support Vector Machine (SVM), Neural Network (NN), or Decision Tree, using extracted mechanical and electrical features as input and lifespan status indicators as output, and learning through training.

[0086] This application addresses the problems of unclear model construction, lack of real-world operating data support, and difficulty in capturing the intrinsic relationship between mechanical and electrical aspects in existing FPC connector life prediction. It proposes a specific method for pre-constructing a coupled correlation model. This method first obtains multiple sets of training samples, providing a solid data foundation for model learning. Each set of training samples is carefully designed, containing not only the mechanical and electrical response signal sequences collected during the testing of an FPC connector sample, but more importantly, the actual life or failure state label corresponding to that FPC connector sample. This three-dimensional combination design of "dual signal sequences + failure labels" enables the model to learn the dynamic evolution of the FPC connector from its initial state to failure, rather than merely the correlation of static data points.

[0087] Mechanical signal sequences reflect the structural fatigue accumulation process of connectors under mechanical stresses such as insertion / removal or vibration, while electrical signal sequences reveal the material degradation and contact performance decline process of the contact interface under mechanical stress. By combining these two dynamic sequences with the final lifetime or failure label, the model can gain a deeper understanding of the intrinsic mechanism by which mechanical fatigue drives electrical performance degradation and ultimately leads to FPC connector failure. Based on this, this method further establishes a correlation equation between the feature values ​​extracted from the mechanical response signal sequence and the feature parameters extracted from the electrical response signal sequence, using these rich training samples. This correlation equation does not simply fit a static relationship between two features, but aims to quantify the dynamic mapping relationship between the mechanical fatigue process and the electrical degradation process, and tightly bind it to the lifetime loss of the FPC connector. For example, when mechanical strain accumulates to a certain level, the charge transfer resistance exhibits a specific changing trend, and this trend has a quantitative relationship with the remaining lifetime of the FPC connector. In this way, the correlation equation can organically combine the "mechanical fatigue process - electrical degradation process - lifetime loss," thus obtaining a coupled correlation model that can accurately characterize the coupling relationship between mechanical and electrical responses. This coupled correlation model is closely integrated with the overall FPC connector lifetime prediction method.

[0088] In actual prediction, once the target FPC connector generates mechanical and electrical response signals during operation, corresponding feature values ​​and parameters can be extracted from these signals and input into a pre-constructed coupled correlation model. Since this model has learned the dynamic evolution of mechanical and electrical characteristics and their mapping relationship with lifespan consumption through the aforementioned method, it can output accurate lifespan prediction results. This synergistic effect ensures that the model can not only identify anomalies in single signals but also capture the coordinated degradation of mechanical and electrical performance during FPC connector failure, thereby significantly improving the accuracy and reliability of lifespan prediction and effectively solving the problems of unclear model training and inability to capture intrinsic correlations in traditional methods.

[0089] As a specific implementation method, the construction process of the pre-built coupling correlation model can be carried out as follows: First, prepare 30 FPC connector samples and divide them into three groups of 10 each. The first group undergoes accelerated insertion and extraction tests under constant insertion and extraction frequency and force; the second group undergoes vibration tests under vibration conditions with specific frequency and amplitude; and the third group undergoes tests under a combined condition of alternating insertion / extraction and vibration. During each test, micro-strain signals at the junction of the flexible substrate and metal terminals of the FPC connector are continuously acquired using micro-strain sensors to form a mechanical response signal sequence. Simultaneously, AC excitation signals are periodically applied through signal electrodes connected to the contact interface of the FPC connector, and their impedance spectrum signals are acquired to form an electrical response signal sequence. When the contact resistance of each FPC connector sample exceeds a preset failure threshold, its total number of insertion / extraction cycles or total operating time is recorded as an actual lifespan label. Subsequently, from the acquired mechanical response signal sequence, feature values ​​such as the maximum strain value and the cumulative strain cycle value within each insertion / extraction cycle can be extracted. From the electrical response signal sequence, feature parameters such as charge transfer resistance and double-layer capacitance can be extracted. Based on these extracted feature values ​​and parameters, along with their corresponding actual lifespan labels, a multivariate nonlinear regression method can be used to establish correlation equations. For example, a multinomial regression model can be constructed, taking mechanical feature values ​​(such as cumulative strain) and electrical feature parameters (such as charge transfer resistance) as inputs, and the lifespan consumption ratio of the FPC connector as the output. By minimizing the error between the predicted and actual lifespans, the model parameters are iteratively optimized, ultimately yielding a coupled correlation model that quantifies the impact of changes in mechanical and electrical characteristics on lifespan. In subsequent practical applications, this model can predict the lifespan status of the FPC connector based on real-time collected mechanical and electrical features.

[0090] Through the above technical solution, this application effectively solves the problems of unclear model construction, lack of real-world operating data support, and difficulty in capturing the intrinsic relationship between mechanical and electrical properties in traditional FPC connector life prediction. Specifically, by acquiring multiple sets of training samples containing mechanical response signal sequences, electrical response signal sequences, and labels of actual life or failure states, the model can learn the dynamic evolution law of the FPC connector from its initial state to failure, rather than merely the correlation of static data points, thus significantly improving the model's adaptability and generalization ability to real-world operating conditions. Based on this, a correlation equation is established between the feature values ​​extracted from the mechanical response signal sequence and the feature parameters extracted from the electrical response signal sequence, realizing a quantitative mapping between the mechanical fatigue process and the electrical degradation process, and tightly binding it to the life consumption of the FPC connector. This method overcomes the limitations of traditional single physical quantity monitoring, enabling the coupled correlation model to more accurately capture the coordinated degradation of mechanical and electrical performance during the failure process of the FPC connector, thereby providing more accurate and reliable life prediction results. Furthermore, this 3D training sample design reduces the model's reliance on massive amounts of data, enabling small and medium-sized enterprises to build high-precision life prediction models without investing significant resources. This effectively reduces implementation costs and provides a clear basis for subsequent model calibration and mass production consistency assurance, ensuring the stability and reliability of the prediction solution in practical applications. Ultimately, this solution provides solid technical support for predictive maintenance of FPC connectors, avoiding equipment downtime losses caused by sudden connector failures.

[0091] In some of the embodiments described above in this application, it is proposed to establish correlation equations to construct a coupled correlation model. However, in the process of its implementation, there is uncertainty in how to efficiently and accurately establish the correlation between the characteristic values ​​of mechanical response signals and the characteristic parameters of electrical response signals. This may lead to insufficient model accuracy, computational complexity, or implementation difficulties, thereby affecting the reliability and practicality of lifetime prediction.

[0092] In this regard, this application further proposes to establish a correlation equation between the feature values ​​extracted from the mechanical response signal sequence and the feature parameters extracted from the electrical response signal sequence, including: using a linear regression method to establish a fitting relationship between the feature values ​​and the feature parameters to obtain the correlation equation.

[0093] Linear regression is a commonly used statistical modeling technique used to analyze the linear relationship between two or more variables. Its core idea is to find a best-fitting straight line (or hyperplane) that minimizes the sum of distances between that line and all data points. In the field of lifespan prediction, linear regression can be used to establish a quantitative relationship between the characteristic values ​​of mechanical response signals and the characteristic parameters of electrical response signals, thereby predicting the lifespan status of FPC connectors.

[0094] As a specific implementation method, specialized statistical analysis software, such as MATLAB, the Scikit-learn library in Python, or the R language, can be used to quickly build models by calling their built-in linear regression functions. Alternatively, algorithms such as the least squares method can be implemented using programming languages ​​(such as C++ and Java) to construct linear regression models. Establishing a fitting relationship between eigenvalues ​​and characteristic parameters refers to using a mathematical model to describe the statistical correlation between the characteristic values ​​of the mechanical response signal (such as maximum strain value, cumulative strain cycle value) and the characteristic parameters of the electrical response signal (such as charge transfer resistance, double-layer capacitance). This fitting relationship aims to capture the inherent connection between the two in the failure evolution process of FPC connectors, thereby enabling the inference of the state of the other based on the change of one. This fitting relationship can be established in various ways. For example, the least squares method can be used to determine the model parameters by minimizing the sum of squared residuals between the predicted and actual observed values, thus obtaining the best-fitting straight line.

[0095] Another approach is to use weighted least squares, assigning different weights to data points based on their reliability or importance to improve fitting accuracy, especially suitable for situations with uneven data quality or outliers. Obtaining the correlation equation means that through the above fitting process, a clear mathematical expression is ultimately formed, which quantifies the relationship between the characteristic values ​​of the mechanical response signal and the characteristic parameters of the electrical response signal. This correlation equation is the core component of the coupled correlation model; it serves as input for subsequent calculations of the FPC connector's lifespan status indicators. For example, if the mechanical response characteristic value is X and the electrical response characteristic parameter is Y, the correlation equation might be in the form Y = aX + b, where a and b are coefficients determined by the fitting process. This equation allows, in actual operation, once the mechanical response signal is acquired and its characteristic values ​​are extracted, the predicted values ​​of the corresponding electrical response characteristic parameters can be directly calculated using this equation, or vice versa, thereby achieving mutual mapping and collaborative analysis of the dual signal characteristics.

[0096] The overall operational logic of this scheme is as follows: First, multiple sets of training samples are pre-acquired. Each set of training samples includes the mechanical response signal sequence and electrical response signal sequence collected during the testing of an FPC connector sample, as well as the actual lifespan or failure state label corresponding to that FPC connector sample. These training samples provide the data foundation for the subsequent establishment of correlation equations. Based on this, the core of this scheme lies in using a linear regression method to fit the feature values ​​extracted from these mechanical response signal sequences with the feature parameters extracted from the electrical response signal sequences. Specifically, the linear regression method describes the quantitative relationship between mechanical feature values ​​and electrical feature parameters by finding an optimal linear function. For example, during the operation of an FPC connector, the accumulation of its mechanical strain (feature value of the mechanical response signal) leads to wear and oxidation of the contact interface (feature parameter of the electrical response signal), thereby increasing the charge transfer resistance. This evolution between mechanical fatigue and electrical degradation often exhibits a significant linear or approximately linear relationship within the core lifespan of the FPC connector.

[0097] Linear regression can accurately capture and quantify this inherent linear correlation. By minimizing the prediction error, it determines the parameters of the fitted line, thus obtaining a correlation equation that accurately reflects the interaction between the two. The establishment of this correlation equation allows for the extraction of corresponding feature values ​​and parameters from the mechanical and electrical response signals generated by the target FPC connector during operation, which are then input into a pre-constructed coupled correlation model during actual life prediction. Since the core correlation equation of this model is based on linear regression, it can efficiently map changes in mechanical characteristics to changes in electrical characteristics, or vice versa, thereby comprehensively evaluating the life status of the FPC connector. This method not only simplifies the complexity of model construction and avoids the overfitting risk that may arise from nonlinear models, but also supports real-time or near-real-time prediction of FPC connector life due to its high computational efficiency. In this way, this solution cleverly combines the linear evolution law of mechanical and electrical characteristics in the failure mechanism of FPC connectors with the advantages of linear regression, ensuring that the constructed coupled correlation model can guarantee prediction accuracy while meeting the requirements of small and medium-sized enterprises for model construction cost and implementation efficiency. It works closely with the steps of acquiring dual signal sequences, extracting feature values ​​and feature parameters to form a complete and efficient FPC connector lifetime prediction system, effectively solving the problems of uncertainty in establishing correlation, insufficient model accuracy and implementation difficulties in traditional methods.

[0098] As a specific implementation method, the correlation equation between the feature values ​​extracted from the mechanical response signal sequence and the feature parameters extracted from the electrical response signal sequence can be established as follows: First, a batch of FPC connector samples can be collected and accelerated aging experiments can be conducted on a test bench simulating actual working conditions. During the experiment, micro-strain signals at the junction of the flexible substrate and metal terminals of the FPC connector are continuously collected using micro-strain sensors, and AC excitation signals are simultaneously applied to the contact interface through signal electrodes and their impedance spectrum signals are collected. At different aging stages, the maximum strain value and the cumulative strain cycle value are extracted from the micro-strain signal as mechanical response feature values, and the charge transfer resistance and double-layer capacitance are extracted from the impedance spectrum signal as electrical response feature parameters. Simultaneously, the actual failure time of each sample or the state label when the preset failure criterion is reached is recorded. Subsequently, these collected data are organized into a training dataset. For example, the cumulative strain cycle value can be selected as the mechanical response feature value X, and the charge transfer resistance as the electrical response feature parameter Y. Using the Python programming language, the LinearRegression module in the scikit-learn library can be imported. By calling LinearRegression().fit(X... train ,Y train The function, where X train It is an array of cyclic cumulative strain values ​​of the training samples, Y train Given the corresponding charge transfer resistance array, a linear regression model can be trained. This model will automatically calculate the optimal slope (regression coefficients) and intercept, resulting in a correlation equation of the form Y=aX+b. For example, if the fitted correlation equation is: Charge transfer resistance = 0.05 * cumulative strain cycle value + 10, this means that for every unit increase in the cumulative strain cycle value, the charge transfer resistance will increase by an average of 0.05Ω. This equation can then serve as the core part of the coupled correlation model, used in actual operation to predict the charge transfer resistance of the FPC connector based on the real-time monitored cumulative strain cycle value, thereby assessing its lifespan.

[0099] Through the above technical solution, this application effectively solves the uncertainty problem of how to efficiently and accurately establish the correlation between the characteristic values ​​of mechanical response signals and the characteristic parameters of electrical response signals when constructing coupled correlation models. This avoids insufficient model accuracy, computational complexity, or implementation difficulties, thus significantly improving the reliability and practicality of life prediction. Specifically, by using linear regression to establish the fitting relationship between characteristic values ​​and characteristic parameters, the significant linear correlation between mechanical fatigue and electrical degradation in the failure evolution process of FPC connectors can be accurately captured, thereby ensuring the accuracy of the life prediction model. Compared to complex nonlinear algorithms, the linear regression method has a clear principle and is computationally simple, significantly lowering the threshold for model construction. This allows small and medium-sized enterprises to quickly build and deploy life prediction models without investing large amounts of resources, effectively meeting their needs for limited resources and rapid implementation. Furthermore, the correlation equation obtained through linear regression has good interpretability; its coefficients can intuitively reflect the influence weight of mechanical characteristics on electrical degradation, facilitating researchers to deeply understand the failure mechanism and perform targeted optimization.

[0100] Meanwhile, the model is easy to calibrate, and parameters can be quickly adjusted with a small amount of new data, ensuring its applicability to different batches of products and operating conditions. Its efficient computational characteristics also enable real-time output of lifespan status indicators. Combined with dynamic signal acquisition, this strongly supports the real-time monitoring and predictive maintenance needs of FPC connector lifespan, avoiding untimely maintenance due to computational lag. Therefore, this solution significantly improves the efficiency of model construction, reduces implementation costs, and enhances the interpretability and maintainability of the model while ensuring prediction accuracy, providing a highly accurate and practical solution for FPC connector lifespan prediction. Example

[0101] Based on Example 1, this embodiment proposes a life status index to predict the life status of FPC connectors. However, in this process, there is a lack of an automated early warning mechanism to trigger maintenance actions in a timely manner, which makes predictive maintenance impossible.

[0102] In this regard, such as Figure 2 As shown, this application further proposes: S300: Compare the lifespan status indicator with a preset warning threshold; when the lifespan status indicator reaches or exceeds the preset warning threshold, generate predictive maintenance prompt information.

[0103] The lifespan status index is a numerical value that quantifies the current health status or remaining service life of an FPC connector. It can be a percentage value representing the health level of the FPC connector, for example, 100% represents brand new and 0% represents complete failure; it can also be a predicted remaining lifespan, such as remaining operating hours; or a dimensionless health index based on a specific model. The purpose of this index is to integrate the characteristic changes of complex mechanical and electrical response signals into a single, easily understood and judged value, thereby providing a basis for subsequent lifespan assessment and maintenance decisions. The preset warning threshold is a critical value used to determine whether an FPC connector needs maintenance or replacement. It can be a fixed value, for example, triggering a warning when the lifespan status index falls below a certain percentage; or it can be a dynamically adjusted value based on the specific application scenario, importance, maintenance cost, historical failure data, or expert experience of the FPC connector. For example, for FPC connectors in critical equipment, the warning threshold can be set higher to ensure earlier maintenance and avoid potential downtime losses; while for non-critical equipment, the threshold can be set relatively lower to optimize maintenance resources. The comparison step aims to assess, through logical judgment, whether the current lifespan status of the FPC connector has reached a point where maintenance measures are required. The comparison operation can be based on simple numerical judgments, such as determining whether the lifespan status indicator is less than or equal to a preset warning threshold (where a lower indicator value indicates a worse condition), or whether it is greater than or equal to a preset warning threshold (where a higher indicator value indicates a worse condition).

[0104] Furthermore, comparisons can also be based on interval judgments, for example, triggering a warning when the lifespan status indicator falls into a certain warning interval. Predictive maintenance alerts are notifications automatically sent to relevant personnel or the system when the FPC connector's lifespan status detects that it has reached a warning condition. These alerts can be implemented in various forms, such as sending text messages via email, SMS, or instant messaging; displaying audible and visual alarms on the monitoring interface or equipment control panel; automatically generating maintenance work orders and sending them to the maintenance management system; or triggering other automated maintenance processes through API interfaces. The purpose of generating these alerts is to transform predictive results into executable maintenance instructions, thereby achieving proactive and preventative maintenance.

[0105] This application's solution constructs an automated closed loop from lifespan prediction to maintenance execution by comparing the lifespan status indicators of the FPC connector with preset warning thresholds in real time and automatically generating predictive maintenance prompts. Specifically, the system first uses a pre-built coupled correlation model to continuously output its lifespan status indicators based on the mechanical and electrical response signals generated by the FPC connector during operation. These lifespan status indicators comprehensively reflect the accumulation of physical damage and the degree of electrical performance degradation within the FPC connector. Subsequently, the system logically compares this real-time updated lifespan status indicator with the preset warning thresholds. These warning thresholds are critical values ​​determined comprehensively based on factors such as the actual application scenario, reliability requirements, and maintenance strategies of the FPC connector. Once the lifespan status indicator reaches or exceeds the warning threshold, it indicates that the FPC connector has entered a stage requiring attention or maintenance measures, and the system immediately and automatically triggers the generation of predictive maintenance prompts. This mechanism ensures that potential failure risks can be identified in a timely manner before actual failure occurs in the FPC connector, and that these risks are transformed into explicit maintenance instructions. In this way, the solution closely integrates the accurate life prediction results in the early stage with the maintenance actions in the later stage, realizing continuous monitoring of the health status of FPC connectors, automatic risk assessment and timely maintenance response. It effectively solves the problem of lack of automated early warning and maintenance triggering mechanisms in traditional solutions, and makes predictive maintenance move from theory to practical application.

[0106] As a specific implementation method, the lifespan status index of the FPC connector can be set as a health index, with a value ranging from 0 to 100, where 100 represents brand new and 0 represents complete failure. A preset warning threshold can be set according to the importance of the FPC connector in a specific application scenario. For example, for FPC connectors in critical industrial equipment, the warning threshold can be set to 30; while for FPC connectors in consumer electronics products, it can be set to 20. During actual operation, the system continuously calculates and updates the health index of the FPC connector. When the health index drops to 30 or below, the system will immediately trigger a warning. At this time, various forms of predictive maintenance prompts can be generated. For example, a maintenance work order can be automatically created through the enterprise's internal maintenance management system and assigned to the corresponding maintenance engineer; simultaneously, an SMS notification can be sent to the equipment operator's terminal device, informing them of the potential failure risk of the FPC connector and suggesting inspection or replacement; in addition, the corresponding FPC connector icon color can be changed from green to yellow or red on the FPC connector status display interface of the central monitoring platform, accompanied by an audible alarm, to attract the attention of management personnel.

[0107] Through the above technical solution, this application effectively solves the problems of existing solutions lacking an automated early warning mechanism and being unable to achieve predictive maintenance. This solution compares the lifespan status indicators of the FPC connector with preset early warning thresholds, enabling real-time and accurate determination of whether the FPC connector has reached a critical state requiring maintenance, thus ensuring timely identification of risks before potential failures occur. When the lifespan status indicators reach or exceed the preset early warning thresholds, the system can automatically generate predictive maintenance prompts, directly converting the prediction results into executable maintenance instructions, avoiding delays caused by manual intervention and errors in subjective judgment. This allows maintenance personnel to proactively inspect, repair, or replace the FPC connector before it actually fails, significantly reducing unexpected equipment downtime, lowering maintenance costs, and improving the overall operational reliability of the equipment. Furthermore, this automated early warning mechanism is closely integrated with the aforementioned FPC connector lifespan prediction method, forming a complete closed loop from data acquisition, feature extraction, model prediction to final maintenance decision-making. This greatly enhances the practicality and operability of the lifespan prediction results, truly realizing predictive maintenance of FPC connectors. Example

[0108] Traditional FPC connector life prediction methods suffer from several drawbacks in applications such as consumer electronics and industrial control for small and medium-sized enterprises. These problems include a disconnect between accelerated aging tests in the laboratory and actual operating conditions, as well as low sensitivity due to monitoring based on a single physical quantity. This leads to either over-design, increasing costs, or underestimating failure risks, resulting in after-sales malfunctions. Furthermore, predictive maintenance is difficult to achieve, causing equipment downtime losses. To address this, this application discloses an electronic device comprising: at least one processor; and a memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform any of the life prediction methods described above.

[0109] The core innovation of this embodiment lies in integrating at least one processor with a memory, enabling the processor to automatically execute a pre-built coupled correlation model to process mechanical and electrical response signals. This solves the problems of disconnect between laboratory testing and actual operating conditions, as well as the low sensitivity of monitoring single physical quantities, achieving the goal of providing a low-cost, high-precision life prediction solution for small and medium-sized enterprises. Specifically, this electronic device serves as a platform, providing a physical carrier for the life prediction method and ensuring that the solution can be practically applied in industrial settings. When at least one processor executes instructions, it processes mechanical and electrical response signal data, applies the pre-built coupled correlation model for calculation, efficiently fuses multi-source signal features, and generates life prediction results, significantly improving the sensitivity to identify early, weak failure signals. The memory stores instructions, saves method steps and running data, ensuring that the prediction process can be repeatedly executed and persistently deployed. When instructions are executed by the processor, it automatically completes steps such as signal acquisition, feature extraction, and model calculation, dynamically responding to changes in the actual operating environment, achieving accurate mapping from operating condition signals to life state, and effectively overcoming the limitations of disconnect between laboratory testing and actual conditions.

[0110] Through the above technical solution, this electronic device can generate predictive maintenance prompts in real time, identify FPC connector failure risks in advance, and reduce downtime losses. Compared with traditional methods that rely on laboratory testing or monitoring of single physical quantities, this solution, through deep integration of hardware and software, provides a solution for SMEs lacking complex operating condition simulation capabilities and big data analysis platforms without additional high-cost investment. It avoids the increased costs caused by over-design and solves the passive maintenance mode of discovering problems only after failure occurs, effectively meeting the ever-increasing reliability requirements of end products for FPC connectors. Example

[0111] Traditional FPC connector life prediction methods have several drawbacks in applications for small and medium-sized enterprises, such as consumer electronics and industrial control. These problems include a disconnect between laboratory accelerated aging tests and actual operating conditions, as well as low sensitivity of monitoring methods based on a single physical quantity. As a result, companies often struggle to accurately assess the lifespan of products in real-world environments. This can lead to increased costs due to over-design or underestimation of failure risks, resulting in after-sales malfunctions. Furthermore, the methods fail to identify early, weak failure signals in a timely manner to enable predictive maintenance, causing equipment downtime losses.

[0112] In response, this application also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method described above. The core innovation of this embodiment lies in using a computer-readable storage medium as a physical carrier to store and distribute the program code implementing the lifetime prediction algorithm, thereby enabling the coupled correlation model to perform automated analysis based on mechanical and electrical response signals during actual operation. Since this model characterizes the intrinsic correlation between changes in mechanical and electrical response signal features, it can more comprehensively capture the early degradation characteristics of FPC connectors, avoiding the limitations of monitoring a single physical quantity, and ensuring that the prediction results closely match the real-world usage environment, thus solving the problem of discrepancies between laboratory test data and actual operating conditions.

[0113] Through the above technical solutions, SMEs can deploy high-precision life prediction systems at low cost without complex operating condition simulation equipment or big data analysis platforms. Specifically, the computer-readable storage medium facilitates program storage and distribution, significantly reducing the pressure on enterprise R&D budgets; during program execution, it automatically processes real-time acquired mechanical and electrical response signals and outputs life prediction results using a pre-built coupled correlation model, realizing real-time monitoring and early warning of the health status of FPC connectors. For example, in industrial control equipment, when the cumulative effect of vibration signals and small changes in contact voltage drop show a synergistic deterioration trend, the model can identify this correlation feature and issue maintenance prompts, thereby effectively supporting predictive maintenance, reducing unexpected equipment downtime losses, and improving the accuracy and foresight of product reliability management.

[0114] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for predicting the lifespan of an FPC connector, characterized in that, include: Acquire the mechanical and electrical response signals generated by the target FPC connector during operation; Based on the mechanical response signal and the electrical response signal, a pre-constructed coupling correlation model is used to determine the lifetime prediction result of the target FPC connector, wherein the coupling correlation model characterizes the correlation between the characteristic changes of the mechanical response signal and the characteristic changes of the electrical response signal.

2. The method as described in claim 1, characterized in that, The mechanical response signal is a micro-strain signal; The acquisition of the mechanical response signal generated by the target FPC connector during operation includes: Micro-strain sensors are used to collect micro-strain signals of the FPC connector under insertion, removal, or vibration conditions by means of micro-strain sensors located at the junction of the flexible substrate and the metal terminal of the FPC connector.

3. The method as described in claim 1, characterized in that, The electrical response signal is an impedance spectrum signal; the acquisition of the electrical response signal generated by the target FPC connector during operation includes: An AC excitation signal is applied to the contact interface by means of a signal electrode connected to the contact interface of the FPC connector, and its impedance spectrum signal is acquired.

4. The method as described in claim 1, characterized in that, The step of determining the lifetime prediction result of the target FPC connector based on the mechanical response signal and the electrical response signal using a pre-built coupling correlation model includes: Extract a first feature value from the mechanical response signal and extract a second feature parameter from the electrical response signal; By inputting the first feature value and the second feature parameter into the coupling correlation model, the lifetime status index of the FPC connector is obtained.

5. The method as described in claim 4, characterized in that, The first characteristic value includes at least one of the following: maximum strain value, strain cycle accumulation value; The second characteristic parameter includes at least one of the following: charge transfer resistance and double-layer capacitance.

6. The method as described in claim 1, characterized in that, The method for constructing the pre-built coupling and correlation model includes: Multiple sets of training samples are obtained. Each set of training samples includes: a mechanical response signal sequence and an electrical response signal sequence collected during the testing of an FPC connector sample, as well as the actual lifespan or failure state label corresponding to the FPC connector sample. Based on the training samples, a correlation equation is established between the feature values ​​extracted from the mechanical response signal sequence and the feature parameters extracted from the electrical response signal sequence to obtain the coupled correlation model.

7. The method as described in claim 6, characterized in that, The process of establishing the correlation equation between the feature values ​​extracted from the mechanical response signal sequence and the feature parameters extracted from the electrical response signal sequence includes: A linear regression method is used to establish a fitting relationship between the feature values ​​and the feature parameters to obtain the correlation equation.

8. The method as described in claim 4 or 7, characterized in that, Also includes: The lifespan status index is compared with a preset early warning threshold; When the lifespan status indicator reaches or exceeds the preset warning threshold, a predictive maintenance prompt message is generated.

9. An electronic device, characterized in that, include: At least one processor; as well as A memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform the method as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 8.

Images