Method for predicting lifetime of fiber optic pressure sensor based on multi-factor aging

Abstract

The application discloses a kind of optical fiber pressure sensor life prediction methods based on multi-factor aging.The method is oriented to interface failure mechanism, assesses the life of sensor, first by standardized test determines the initial shear strength of adhesive layer material;Then predict working shear stress by finite element analysis;Then design and execute multi-factor accelerated aging test, obtain the strength degradation data under the joint action of temperature, humidity, cyclic loading, finally combine real-time monitoring data to correct working shear stress, calculate the service life under specified reliability based on Weibull statistics.The application first improves Eyring model, and combines the improved Eyring model with strength degradation model, establishes a systematic parameter determination standard and reliability evaluation system, significantly improves the reliability and service life predictability of sensor in actual engineering.

Description

Technical Field

[0001] This invention belongs to the fields of fiber optic sensing and intelligent detection and civil engineering technology, specifically a method for predicting the lifespan of a fiber optic pressure sensor based on multi-factor aging. Background Technology

[0002] Fiber optic sensors play an irreplaceable role in long-term health monitoring in fields such as civil engineering, aerospace, and energy and power due to their unique advantages, including resistance to electromagnetic interference and the ability to achieve long-distance, long-term monitoring. However, in practical engineering applications, sensors are often buried or installed in complex and harsh environments such as concrete and soil, and are subjected to the coupling effects of multiple factors such as temperature, humidity, chemical media, and cyclic loads over a long period of time.

[0003] The complex service environment causes sensor performance to gradually degrade over time. The most common and critical failure mode is the debonding of the adhesive layer of the sensing element or interface failure. Under the combined effects of heat, humidity, and stress, the adhesive material fixing the sensing element undergoes hydrolysis, plasticization, creep, and fatigue damage, leading to a continuous decrease in its adhesive strength. Ultimately, this causes the sensor to detach from the substrate, resulting in the loss of monitoring functionality. This failure process is insidious and gradual, making it difficult to detect in the field, yet its consequences are severe: it can lead to the loss of the most valuable monitoring data when critical changes occur in the safety status of the engineering structure, resulting in significant safety hazards and incalculable economic losses.

[0004] Currently, there is a lack of systematic and reliable methods for lifetime assessment of sensors, especially fiber optic sensors. Traditional lifetime predictions are mostly based on a single accelerated aging factor (such as high temperature or high humidity), failing to fully consider the synergistic effects of multiple coupled factors in actual operating conditions, leading to significant deviations between predicted and actual lifetimes. Furthermore, existing methods focus on the degradation of the material itself or the overall degradation of the sensor, failing to establish a complete correlation model from material performance degradation to sensor system functional failure starting from the failure mechanism of the key mechanical transmission path of the "binder-interface".

[0005] Patent CN110162860A uses the entire strain measurement system as the research object and the percentage change in the relative strain measurement value of the system's final output as the evaluation index for performance degradation. It achieves long-term lifespan prediction through a relatively short-term temperature-accelerated experiment. However, this approach is too simplistic in its accelerated aging factors, considering only temperature as the accelerating stress and ignoring complex actual operating conditions. It employs a "black box" strategy, avoiding analysis of key interfaces, resulting in an empirical fit with insufficient physical meaning. This patent application, on the other hand, does not use the system's final output parameters as the sole evaluation criterion. Instead, it addresses the failure of key interfaces affecting lifespan by comprehensively incorporating multiple factors such as temperature, humidity, and cyclic loading to establish a lifespan prediction method based on failure mechanisms. This method offers advantages such as more accurate consideration of multiple factors compared to actual operating conditions, more precise characterization of key failure behaviors, a more comprehensive physical basis for the model, and higher prediction accuracy and engineering applicability. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention studies a fiber optic pressure sensor lifetime prediction method based on multi-factor aging. The aim is to use interface failure mechanism as a guide, take the lifetime of the sensor's key interface as the evaluation object, and combine humidity, cyclic load, and temperature to form accelerated stress. By analyzing the lifetime problem from a mechanistic perspective, and facing the complex multi-factor environment on the ground, a more systematic, refined, and reliable multi-factor lifetime prediction system is proposed.

[0007] The specific solution of the present invention is as follows: S1. Single-lap shear specimens were prepared using the same substrate and bonding material as the sensor. The thickness of the bonding layer was controlled to be 0.1±0.02mm. Tensile tests were conducted and the maximum failure load was recorded. The shear strength of the bonding layer was calculated accordingly. The obtained shear strength data were subjected to Weibull statistical analysis, and the 5th percentile value was selected as the initial shear strength of the bonding layer.

[0008] S2. Establish a refined finite element model of the sensor's elastic element and conversion element, refine the mesh of the bonding layer region, apply constraints and loads according to the actual working conditions, extract the shear stress distribution of the bonding layer region, and take the maximum shear stress value as the expected working shear stress.

[0009] S3. Accelerated aging tests were conducted using a three-factor, two-level full-factor experimental design. The selected factors included temperature, humidity, and cyclic loading. The temperature was set at 50℃ and 80℃, the humidity was set at 50%RH and 85%RH, and the cyclic loading was set with a threshold of 30% of the initial shear strength. Two states were set with and without cyclic loading. The degradation data of the shear strength of the adhesive layer under different combinations of conditions were recorded.

[0010] S4. For each accelerated aging condition, the relationship between the degradation time constant and temperature, humidity and load conditions is established using a multiple linear regression method (modified Eyring model), and a shear strength degradation model is integrated to establish a fiber optic sensor lifetime prediction model. In the formula, This refers to the failure time (lifetime). The theoretical degradation time constant, For activation energy, Boltzmann's constant, Absolute temperature The humidity acceleration coefficient, Relative humidity, The load acceleration factor, For loading factor, The initial shear strength, This is the working shear stress.

[0011] S5. In actual use environments, the model parameters are corrected based on the environmental parameters obtained from monitoring, combined with Weibull statistical analysis, thereby calculating the service life under a specified reliability: In the formula, The specified reliability (maximum 1). The shape parameter is the Weibull distribution. Furthermore, at least 30 single-lap shear specimens are prepared in S1. This increases the stability and reliability of the statistical analysis and reduces the impact of random errors on the initial strength estimation. After the test, the measured shear strengths are arranged in ascending order, and the failure probability corresponding to each strength value is calculated using the median rank formula. The shape parameter is then calculated using Weibull statistical analysis. Scale parameters .

[0012] Furthermore, in S3, a preliminary experiment on the cyclic loading frequency was conducted: ensuring the total number of loading cycles was the same, the specimens were cyclically loaded at different frequencies, and the load-displacement curves and temperature changes were monitored. The experimental results were analyzed, and a loading frequency was selected where the surface temperature of the specimen increased by no more than 1°C during loading, and the lifespan did not change significantly with increasing frequency. This avoids the coupling of thermal aging effects and force effects, ensuring that the strength degradation observed in subsequent accelerated aging tests is purely due to mechanical fatigue caused by the "cyclic loading" factor. Further, in S4, the specific steps for establishing the relationship between the degradation time constant and temperature, humidity, and load conditions were as follows: first, the logarithm of the modified Eyring model was taken to obtain its linear form. Secondly, the simplified formula is: Next, each condition of As the independent variable, Using the dependent variable, a multiple linear regression was performed, yielding... , , , Finally, the model coefficients are derived by reverse calculation.

[0013] Furthermore, model coefficients obtained from each experiment can be collected to establish a database of acceleration coefficients for different interfaces.

[0014] Furthermore, the shape parameters in the reliable lifespan of S5 The average value under different acceleration conditions is taken to reduce the impact of random error in a single experiment on the statistical parameters. Attached Figure Description

[0015] The invention will now be further described with reference to the accompanying drawings.

[0016] Figure 1 This is a flowchart of a fiber optic pressure sensor lifetime prediction method based on multi-factor aging according to the present invention. Detailed Implementation

[0017] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The described embodiments are only some embodiments of this application, and not all embodiments.

[0018] This invention proposes a method for predicting the lifetime of fiber optic pressure sensors based on multi-factor aging. Please refer to [link / reference]. Figure 1 The detailed steps of this method are shown below: S1. Prepare single-lap shear specimens using the same metal and bonding materials as the sensor. The number of specimens should be at least 30. A sufficient sample size is necessary to allow for the determination of shape parameters in Weibull statistical analysis. Relatively stable, resulting in more reliable life prediction; a 0.1±0.02mm metal wire was placed at the edge of the bonding area to control the adhesive layer thickness within the range of 0.1±0.02mm, ensuring uniform stress distribution and mechanical property stability of the specimen. Tensile tests were conducted at a loading rate of 1.27mm / min in an environment of 24℃ and 50%RH, recording the maximum failure load for each group of specimens and calculating the corresponding shear strength; the measured shear strengths were arranged in ascending order, and the median rank formula was used. , For serial numbers (from 1 to ), calculate the failure probability estimate for each intensity; and perform a cumulative distribution function on the Weibull distribution. In the formula, For the first Estimated failure probability corresponding to each shear strength For the first Shear strength.

[0019] A linear fit was performed on the thirty data points to obtain... and The estimated value, It can be used as a quantitative index of the dispersion of bond strength, and then utilized Calculate the initial shear strength at 95% reliability.

[0020] S2. Establish a refined finite element model of the sensor's elastic element and conversion element, refine the mesh of the adhesive layer region, apply constraints and loads according to the actual working conditions, extract the shear stress distribution of the adhesive layer region, take the maximum shear stress value as the expected working shear stress, and use this to design the cyclic load threshold in the aging test.

[0021] S3. A three-factor, two-level full-factor experimental design was used to conduct accelerated aging tests, with temperatures set at 50℃ and 80℃. 50℃ is higher than the sensor's maximum operating temperature to ensure a significant accelerated aging effect; 80℃ is lower than the glass transition temperature of the adhesive layer material to avoid introducing new failure mechanisms due to material decomposition. Simultaneously, the temperature difference between 50℃ and 80℃ is sufficiently large to produce a significant difference in aging rates. Humidity was set at 50%RH and 85%RH. 50%RH represents the average humidity of the normal operating environment, while 85%RH is used to significantly accelerate the diffusion and penetration of moisture in the material. For simulations of extremely humid environments, the humidity can be increased to a maximum of 95%RH. Cyclic loading used 30% of the initial shear strength as the load threshold, with and without cyclic loading. Before applying cyclic loading, the loading frequency needed to be determined through pre-experiments: this frequency should ensure that, with a constant total number of loading cycles, an increase in frequency does not significantly change the sample life, and the surface temperature rise of the sample during loading does not exceed 1℃. For each acceleration condition, samples were taken at time intervals of 0, 12, 24, 96, 240, and 480 to measure shear strength. The approximately logarithmic time intervals better match the degradation curve of the material. In the early rapid aging stage, denser sampling points were set to quickly understand the trend and rate of strength degradation.

[0022] S4. Improved Eyring model considering temperature, humidity and load conditions ( (Take 1 when there is a load, and 0 when there is no load).

[0023] Taking the natural logarithm of the model yields a linear form:

[0024] Simplify the formula variables:

[0025] Next, each condition of As the independent variable, Using the dependent variable, a multiple linear regression was performed, yielding... , , , Finally, the model coefficients are derived to establish a fiber optic sensor lifetime prediction model: In the formula, This refers to the failure time (lifetime). The theoretical degradation time constant, For activation energy, Boltzmann's constant, Absolute temperature The humidity acceleration coefficient, Relative humidity, The load acceleration factor, For loading factor, The initial shear strength, This is the working shear stress.

[0026] Acceleration coefficient databases were established based on experimentally obtained acceleration coefficients for different bonding materials to facilitate the use of subsequent lifetime models.

[0027] In actual monitoring, based on the differential principle, the shear stress is derived by calculating the shear strain using the differential signal, the actual temperature is calculated using the common-mode signal, and the service life is calculated by substituting these values ​​into the model. The service life is then corrected using Weibull statistical analysis to calculate the service life under a specified reliability. In the formula, The specified reliability (maximum 1). The shape parameter of the Weibull distribution ( (Take the average value under different acceleration conditions).

Claims

1. A method for predicting the lifetime of an optical fiber pressure sensor based on multi-factor aging, characterized in that, include: S1. Single-lap shear specimens were prepared using the same substrate and adhesive as the sensor, with the adhesive thickness controlled at 0.1±0.02 mm. Tensile tests were conducted and the maximum failure load was recorded. The shear strength of the adhesive layer was calculated based on this. Weibull statistical analysis was performed on the obtained shear strength data, and the 5th percentile value was selected as the initial shear strength of the adhesive layer. S2. Establish a refined finite element model of the sensor's elastic element and conversion element, refine the mesh of the bonding layer region, apply constraints and loads according to the actual working conditions, extract the shear stress distribution of the bonding layer region, and take the maximum shear stress value as the expected working shear stress. S3. Accelerated aging tests were conducted using a three-factor, two-level full-factor experimental design. The selected factors included temperature, humidity, and cyclic loading. The temperature was set at 50℃ and 80℃, the humidity was set at 50%RH and 85%RH, and the cyclic loading was set with a threshold of 30% of the initial shear strength. Two states were set with and without cyclic loading. The degradation data of the shear strength of the adhesive layer under different combinations of conditions were recorded. S4. For each accelerated aging condition, the relationship between the degradation time constant and temperature, humidity and load conditions is established using a multiple linear regression method (modified Eyring model), and a shear strength degradation model is integrated to establish a fiber optic sensor lifetime prediction model. In the formula, This refers to the failure time (lifetime). The theoretical degradation time constant, For activation energy, Boltzmann's constant, Absolute temperature The humidity acceleration coefficient, Relative humidity, The load acceleration factor, For loading factor, The initial shear strength, This is the working shear stress. S5. In actual use environments, the model parameters are corrected based on the environmental parameters obtained from monitoring, combined with Weibull statistical analysis, thereby calculating the service life under a specified reliability: In the formula, The specified reliability (maximum 1). Let be the shape parameter of the Weibull distribution.

2. The method for predicting the lifetime of an optical fiber pressure sensor based on multi-factor aging as described in claim 1, characterized in that, The number of single-lap shear specimens prepared in S1 is at least 30.

3. The method for predicting the lifetime of an optical fiber pressure sensor based on multi-factor aging as described in claim 1, characterized in that, In step S1, the shear strengths measured by the single lap shear test are arranged in ascending order, and the failure probability corresponding to each strength value is calculated using the median rank formula. This is then used for Weibull statistical analysis to calculate shape parameters. Scale parameters .

4. The method for predicting the lifetime of an optical fiber pressure sensor based on multi-factor aging as described in claim 1, characterized in that, The loading frequency of the cyclic load in step S3 is determined through preliminary experiments. Specifically, the total number of loading cycles is kept the same, and the sample is cyclically loaded at different frequencies to monitor its load-displacement curve and surface temperature changes. Based on the criteria that the surface temperature rise of the sample does not exceed 1°C and the sample life does not change significantly when the frequency increases, the final loading frequency is selected from the frequencies.

5. The method for predicting the lifetime of an optical fiber pressure sensor based on multi-factor aging according to claim 1, characterized in that, In S3, for each accelerated aging condition, samples are taken out at time intervals of 0, 12, 24, 96, 240, and 480 hours to determine the shear strength.

6. The method for predicting the lifetime of an optical fiber pressure sensor based on multi-factor aging according to claim 1, characterized in that, The specific steps in S4 to establish the relationship between the degradation time constant and temperature, humidity, and load conditions are as follows: Take the natural logarithm of the improved Eyring model to obtain a linear form. Simplify the formula to Each condition of As the independent variable, Using the variable as the dependent variable, perform multiple linear regression to obtain the corresponding... , , , The model coefficients are then derived by reverse calculation.

7. The method for predicting the lifetime of an optical fiber pressure sensor based on multi-factor aging according to claim 1, characterized in that, In S5, the shape parameters of the Weibull analysis Take the average value under different acceleration conditions.

8. The method for predicting the lifetime of an optical fiber pressure sensor based on multi-factor aging as described in claim 6, characterized in that, The acceleration coefficients derived from the experiments were used to establish an acceleration coefficient database for different interfaces.

Images