A method for predicting healing performance of asphalt material based on coupling effect of damage and time
By using cyclic experiments and nonlinear model fitting, a method for predicting the healing performance of asphalt materials coupled with damage and time was established. This method solves the problem of the lack of coupling between damage state and healing time in the existing technology, and realizes dynamic prediction and maintenance optimization of the healing performance of asphalt materials.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- 金华市金东区公路与运输管理中心
- Filing Date
- 2026-03-16
- Publication Date
- 2026-07-14
AI Technical Summary
Existing technologies fail to effectively combine the two variables of damage state and healing time, making it impossible to predict the dynamic evolution of the healing performance of asphalt materials, resulting in biased fatigue life prediction and unreasonable maintenance timing.
Through cyclic tests of fatigue loading, intermittent healing, and re-fatigue loading, the initial damage density and healing index were obtained. Feature parameters were extracted using a nonlinear model fitting, a quantitative functional relationship was established, a comprehensive mathematical model was constructed, and damage density and healing time were coupled to predict the dynamic evolution of the healing index under any damage state.
It enables dynamic prediction of the healing performance of asphalt materials, improves the accuracy of fatigue life prediction, optimizes maintenance decisions, avoids resource waste and structural damage, and provides a scientific basis for pavement management.
Smart Images
Figure CN122392685A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of road engineering materials technology, and in particular to a method for predicting the healing performance of asphalt materials based on the damage-time coupling effect. Background Technology
[0002] With the rapid development of highway transportation, asphalt pavement has been widely used due to its excellent driving smoothness and comfort. However, under long-term repeated loading, asphalt pavement inevitably suffers fatigue damage, specifically manifested as the initiation, propagation, and continuous accumulation and connection of microcracks within the structure. When cracks propagate to a critical level, they lead to a rapid decline in pavement structural strength and performance degradation, ultimately resulting in macroscopic cracking, potholes, and other pavement defects. It is noteworthy that the vehicle load patterns on the road create certain intervals in the stress on the pavement structure. During these intervals, the asphalt binder can undergo a degree of self-healing, i.e., the molecular chains at the crack tips or interfaces re-entangle and reorganize, thereby restoring some mechanical properties. This healing behavior has a significant effect on extending pavement fatigue life, but its positive impact is generally not fully considered in current pavement fatigue life prediction and maintenance decision-making systems. Therefore, scientifically quantifying and predicting the healing behavior of asphalt materials is crucial for achieving accurate prediction of pavement fatigue life and optimized decision-making regarding maintenance timing.
[0003] Existing technologies for evaluating the performance of asphalt materials have been extensively studied, but significant limitations remain. For example, Chinese patent application CN119494217A, entitled "A Method for Predicting the Fatigue Life of Recycled Asphalt Based on Activation Energy," focuses on predicting the fatigue life of recycled asphalt by analyzing the difference in aging activation energy between new and old asphalt. While this method quantifies the impact of old asphalt incorporation from an aging kinetics perspective, its technical solution entirely emphasizes the aggravating effect of long-term chemical oxidation during aging on fatigue damage, without addressing the physicochemical process of self-healing during load breaks—a process that can restore damage. Therefore, this prediction model is essentially still a model that only considers the unidirectional accumulation of damage, failing to reflect and utilize the inherent self-healing ability of the material. Its prediction results may tend to be conservative, leading to an underestimation of the remaining pavement life.
[0004] On the other hand, for the direct testing and evaluation of asphalt self-healing performance, existing technologies, such as the Chinese invention patent application CN113433304A entitled "Testing and Evaluation Method of Asphalt Self-Healing Performance," provide a method for comprehensively evaluating the recovery of asphalt modulus and fatigue resistance performance through fatigue-intermittent-fatigue tests and the introduction of a complex healing index. This method provides an important indicator for the comparative evaluation of healing performance. However, it is essentially a static performance testing and comparison method conducted under specific preset damage conditions and specific fixed healing time conditions. The healing index obtained by this method is a result value under specific test conditions, rather than a process function. It lacks a description of the dynamic evolution of healing behavior over time and fails to establish a quantitative correlation between healing capacity and the current damage state of the material. Therefore, this method cannot predict the healing index of the material under different damage degrees and different healing durations, that is, it does not have the "dynamic prediction ability" for arbitrary damage states and arbitrary healing times.
[0005] In summary, the shortcomings of existing technologies lie in the fact that neither fatigue life prediction models focusing on the effects of aging nor static comparative evaluation methods targeting healing have solved a key technical problem: how to establish a quantitative model that can simultaneously couple the two key variables of damage state and healing time to achieve accurate prediction of the dynamic evolution of the healing index of asphalt materials. Without such a model, it is difficult to scientifically incorporate the healing behavior of materials into the pavement damage evolution and life prediction system, thus hindering the full utilization of the self-healing potential of materials in maintenance decisions to achieve resource conservation and life extension. Summary of the Invention
[0006] The purpose of this invention is to overcome the shortcomings of the existing technology by providing a method for predicting the healing performance of asphalt materials based on the coupling effect of damage and time. This method can accurately characterize the dynamic evolution process of asphalt healing and can be combined with damage prediction models in the future to effectively improve the accuracy of pavement fatigue life prediction, providing a solid theoretical foundation and technical support for pavement maintenance optimization decisions.
[0007] The objective of this invention can be achieved through the following technical solutions: This invention provides a method for predicting the healing performance of asphalt materials based on the damage-time coupling effect, comprising the following steps: S1. Through cyclic tests of fatigue loading, intermittent healing, and re-fatigue loading, the initial damage density and healing index of asphalt materials under different initial damage states and different healing times were obtained. S2. Based on the obtained data on the corresponding healing index and healing time, a nonlinear model that can characterize the evolution of the healing process from a rapid stage to a slow stage is used for fitting, thereby extracting multiple key feature parameters that characterize the material's healing behavior. S3. Based on the correspondence between multiple key feature parameters and data of different initial damage states, establish quantitative functional relationships between each feature parameter and the initial damage density to obtain the influence law of damage degree on healing rate and healing potential. S4. Substitute the quantitative functional relationship described in S3 into the nonlinear model described in S2, and couple the two variables of initial damage density and healing time to construct a comprehensive mathematical model that can predict the dynamic evolution of the healing index of asphalt material over time under any damage state. S5. The comprehensive mathematical model is used to predict the healing index of asphalt material under the conditions of initial damage state and healing time.
[0008] Furthermore, in S1, the specific process for obtaining the initial damage density and healing index of asphalt materials under different initial damage states and different healing times includes: The asphalt material sample was subjected to the first sinusoidal oscillatory fatigue loading until the complex shear modulus of the material decayed to the corresponding preset damage density and the loading was stopped. The sample was left to stand at a constant temperature for several pre-set different durations to achieve intermittent healing; A sinusoidal oscillatory fatigue load with the same conditions as the first loading was applied again, and the initial damage density and healing index were calculated based on the initial complex shear modulus of the first loading, the complex shear modulus at the end of the first loading, and the initial complex shear modulus of the second loading.
[0009] Further, in S1, the damage density and healing index are calculated based on the initial complex shear modulus of the first loading, the complex shear modulus at the end of the first loading, and the initial complex shear modulus of the second loading. The specific process includes: The damage density at the end of the first fatigue is calculated based on the ratio of the complex shear modulus at the end of the first loading to the initial complex shear modulus at the beginning of the first loading. Then, the damage density at the end of intermittent healing is calculated based on the ratio of the initial complex shear modulus at the beginning of the second loading to the initial complex shear modulus at the beginning of the first loading. The damage density at the end of the intermittent healing is compared with the damage density at the end of the first fatigue. The healing index, which characterizes the degree of recovery of the material's self-healing ability under the corresponding healing conditions, is obtained by calculating the ratio of the difference between the two to the damage density at the end of the first fatigue.
[0010] Furthermore, in S2, the specific process of fitting the model using a nonlinear model capable of characterizing the evolution of the healing process from a rapid phase to a slow phase includes: The healing index data measured at different healing times were substituted into the Ramberg-Osgood model equation for nonlinear fitting. By judging the goodness of fit, three characteristic parameters in the equation were determined, which respectively characterize the rapid recovery in the early stage of healing, the slow recovery in the later stage of healing, and the maximum recoverable capacity of the material. These correspond to the short-term healing rate parameter, the long-term healing rate parameter, and the healing potential parameter.
[0011] Furthermore, in S2, by determining the goodness of fit, three characteristic parameters in the equation are identified, representing rapid recovery in the early stage of healing, slow recovery in the later stage of healing, and the maximum recoverable capacity of the material, respectively. The specific process includes: After substituting the healing index data at different healing times into the Ramberg-Osgood model equation for nonlinear fitting, the model's coefficient of determination is calculated and evaluated by comparing the differences between the calculated values and the experimental values. When the coefficient of determination is higher than the preset accuracy standard, the fitting result is accepted. The parameters in the fitting equation that characterize the healing rate in the early stage of healing, the healing rate in the later stage of healing, and the maximum healing capacity limit are respectively assigned as the short-term healing rate parameter, the long-term healing rate parameter, and the healing potential parameter.
[0012] Furthermore, in S3, the specific process of establishing the quantitative functional relationship between each characteristic parameter and the initial damage density includes: For the short-term healing rate parameter, long-term healing rate parameter, and healing potential parameter extracted from S2, power function correlation equations were established between them and the initial damage density. Multiple sets of characteristic parameter data under different initial damage densities were used to fit each power function equation to determine the coefficients in each equation, so as to quantify the decay law of each characteristic parameter as the initial damage density increases.
[0013] Furthermore, in S3, multiple sets of characteristic parameter data under different initial damage densities are used to fit each power function equation to determine the coefficients in each equation. The specific process includes: For short-term healing rate parameters, long-term healing rate parameters, and healing potential parameters, their corresponding values at multiple different initial damage density levels were collected. With initial damage density as the independent variable and each characteristic parameter as the dependent variable, a nonlinear regression analysis was performed on the data using power function equations. By maximizing the coefficient of determination between the calculated values of the equations and the measured data points, specific power exponent coefficients and proportional coefficients in each power function equation were obtained. In this way, a precise power function decay relationship between each characteristic parameter and the initial damage density was established.
[0014] Furthermore, in S4, by coupling the initial damage density and healing time, a comprehensive mathematical model is constructed that can predict the dynamic evolution of the healing index of asphalt materials over time under any damage state. The specific process includes: The three power function relationships obtained in S3, which characterize the short-term healing rate parameter, the long-term healing rate parameter, and the healing potential parameter, respectively, are related to the initial damage density. These are then substituted into the Ramberg-Osgood model equation used in S2 to obtain an explicit prediction equation for the healing index with the initial damage density and healing time as independent variables, which serves as a comprehensive mathematical model.
[0015] Furthermore, it also includes a model application extension step, which includes: based on the constant reference temperature test data constructed in S4, the comprehensive mathematical model is scaled by introducing a temperature-related shift factor to equivalently scale the healing time variable in the model according to the time-temperature equivalence principle, thereby extending the healing index prediction method to evaluate the healing behavior of asphalt materials under different service temperature conditions.
[0016] Furthermore, in S5, the specific process includes: The initial damage density value and healing time value to be predicted are directly input into the explicit prediction equation corresponding to the comprehensive mathematical model. By substituting the initial damage density value into the calibrated power function relating the characteristic parameters and damage density, the corresponding short-term healing rate parameter value, long-term healing rate parameter value, and healing potential parameter value are calculated. Then, these parameter values, together with the healing time value, are substituted into the Ramberg-Osgood model equation for calculation, and finally, the predicted healing index value under the corresponding conditions is directly output.
[0017] Compared with the prior art, the present invention has the following beneficial effects: 1) At the theoretical level, the method of this invention breaks through the limitations of existing technologies, which can only perform static and comparative evaluations under fixed damage states and healing times. For the first time, it constructs a comprehensive mathematical model that simultaneously couples two core variables: initial damage density and healing time. This model extracts and correlates the influence of damage degree on healing characteristic parameters (such as healing rate and healing potential), achieving accurate prediction of the dynamic evolution of the healing index of asphalt materials under arbitrary damage states over time. This elevates the evaluation of material healing performance from static results to a dynamic process description, providing a key tool for solving the core technical problem of accurately characterizing the dynamic evolution of asphalt healing.
[0018] 2) At the engineering application level, the establishment of the theoretical model of this invention has significant practical value. Because this model can quantitatively predict the amount of material healing and recovery under different damage degrees and interval durations, it makes it possible to scientifically incorporate the self-healing effect of materials into the pavement fatigue life prediction system. By combining this healing prediction model with the pavement damage evolution model, the bias of traditional models that only consider damage accumulation while ignoring healing and recovery can be corrected, thereby significantly improving the prediction accuracy of the remaining fatigue life of the pavement. This provides a solid theoretical basis and technical support for formulating more scientific and economical pavement preventive maintenance decisions, helping to avoid insufficient or excessive maintenance, and ultimately achieving the goals of extending pavement service life and saving maintenance resources. Attached Figure Description
[0019] Figure 1 The effect of healing time and initial lesion density on the healing index; Figure 2 This is a comparison between the model's predicted values and the measured values. Detailed Implementation
[0020] In the conceptualization process of this invention, it was considered that, considering existing technologies, although temperature has a significant impact on the healing behavior of asphalt binders, its influence can be transformed at different temperatures using the time-temperature equivalence principle (existing technology description: Estimating the Healing Characteristic of Asphalt Binder Using the LASHTest). In other words, the healing process at different temperatures can essentially be viewed as a scaling change over a time scale, and its impact can be corrected using a shift factor. Based on this, in the modeling and analysis process of this invention, the temperature effect is treated as a known moderating term that can be corrected through data processing standardization, rather than an independent model variable. Specifically, this invention conducts experiments under constant reference temperature conditions, focusing on developing a method to obtain a quantitative relationship between the healing index, damage density, and healing time. Subsequently, based on the specific needs of engineering analysis, the healing curve can be corrected using a shift factor according to the time-temperature equivalence principle, based on the relationship model at this reference temperature, thereby enabling the prediction of healing behavior under other temperature conditions. Therefore, the focus of this invention is to establish a healing index prediction method applicable to different damage degrees and healing time conditions, while the influence of the temperature effect can be extended through the equivalent transformation process in existing technologies without affecting the main structure and core content of this invention.
[0021] In summary, this invention proposes a method for predicting the healing performance of asphalt materials under multiple time scales and damage levels. This method involves: using a Ramberg-Osgood type equation to describe the change curve of the asphalt binder healing index with healing time under different initial damage densities, thus obtaining the material's healing evolution characteristic equation; further extracting three physically meaningful characteristic parameters for healing rate and healing potential: short-term healing rate, long-term healing rate, and healing potential; establishing quantitative relationships between these three parameters and the initial damage density variable using a sigmoid logistic function; and finally forming a comprehensive healing index prediction model coupled with damage degree and healing time variables. This model can achieve dynamic prediction of the healing index, providing a methodological reference for the prediction and correction of asphalt pavement fatigue life, thereby enabling optimized decision-making on pavement maintenance timing. This avoids resource waste caused by premature maintenance and prevents structural damage caused by delayed maintenance, achieving accurate prediction and management of pavement life considering healing behavior.
[0022] The technical problem to be solved by this invention includes at least one of the following: (1) Fatigue life prediction bias: Most existing fatigue life prediction models for asphalt pavements only take damage accumulation as the main variable and ignore the self-healing effect that asphalt binder may produce during the fatigue load interval. This neglect leads the model to only consider the unidirectional accumulation of damage and does not reflect the partial recovery ability of the material. As a result, the predicted pavement life in actual engineering is often too low or too high, which is significantly different from the actual use condition of the pavement.
[0023] (2) Inappropriate timing of road maintenance: Since the existing fatigue life model does not fully consider the healing effect, two types of deviations are likely to occur in actual maintenance management: On the one hand, some pavements may still have a certain self-healing ability when the model predicts that the life will reach the threshold. Premature maintenance will result in a large waste of resources such as asphalt materials, crushed stone, manpower and energy; on the other hand, if we rely too much on damage accumulation and ignore the healing law, the maintenance time may be delayed, resulting in irreversible expansion of structural damage, thereby increasing long-term maintenance costs and road use risks.
[0024] (3) Lack of quantitative description of the dynamic evolution of healing process: Existing studies on the healing performance of asphalt binders mainly focus on the comparison of material healing indices under fixed test conditions, which is a static evaluation method. Such methods can only reflect the relative healing ability of materials under specific conditions, and cannot reveal the dynamic evolution law of healing rate and healing potential as the damage state changes. It is difficult to provide a quantitative basis for fatigue life prediction and pavement maintenance decision-making.
[0025] (4) The multi-factor relationship between damage and healing is not quantified: In the existing technology, there is a lack of systematic methods to establish a quantitative relationship between the degree of damage and key healing parameters (such as short-term healing rate, long-term healing rate and healing potential). Since different damage levels will significantly affect crack closure, interfacial contact and molecular chain recombination processes, it is impossible to accurately quantify the trend of healing ability with damage, which limits the application of healing models in practical engineering.
[0026] (5) Existing research mostly focuses on experimental analysis at the material level, lacking predictive methods that can couple healing behavior with factors such as damage state and healing time in conjunction with actual pavement usage conditions and are applicable to engineering practice. This limits the application of the healing effect in pavement life assessment and maintenance management, and also fails to provide reliable theoretical support for scientifically determining maintenance timing.
[0027] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Component models, material names, connection structures, circuit structures, control methods, algorithms, and other features not explicitly described in this technical solution are considered common technical features disclosed in the prior art.
[0028] Example 1 The method for predicting the healing performance of asphalt materials based on the damage-time coupling effect in this embodiment includes the following steps: S1. Through cyclic tests of fatigue loading, intermittent healing, and re-fatigue loading, the initial damage density and healing index of asphalt materials under different initial damage states and different healing times were obtained.
[0029] In specific implementation, S1, the specific process of obtaining the initial damage density and healing index of asphalt materials under different initial damage states and different healing times includes: The asphalt material sample was subjected to the first sinusoidal oscillatory fatigue loading until the complex shear modulus of the material decayed to the corresponding preset damage density and the loading was stopped. The sample was left to stand at a constant temperature for several pre-set different durations to achieve intermittent healing; A sinusoidal oscillatory fatigue load with the same conditions as the first loading was applied again, and the initial damage density and healing index were calculated based on the initial complex shear modulus of the first loading, the complex shear modulus at the end of the first loading, and the initial complex shear modulus of the second loading.
[0030] In specific implementation, in S1, the damage density and healing index are calculated based on the initial complex shear modulus of the first loading, the complex shear modulus at the end of the first loading, and the initial complex shear modulus of the second loading. The specific process includes: The damage density at the end of the first fatigue is calculated based on the ratio of the complex shear modulus at the end of the first loading to the initial complex shear modulus at the beginning of the first loading. Then, the damage density at the end of intermittent healing is calculated based on the ratio of the initial complex shear modulus at the beginning of the second loading to the initial complex shear modulus at the beginning of the first loading. The damage density at the end of the intermittent healing is compared with the damage density at the end of the first fatigue. The healing index, which characterizes the degree of recovery of the material's self-healing ability under the corresponding healing conditions, is obtained by calculating the ratio of the difference between the two to the damage density at the end of the first fatigue.
[0031] S2. Based on the obtained data on the corresponding healing index and healing time, a nonlinear model that can characterize the evolution of the healing process from a rapid stage to a slow stage is used for fitting, thereby extracting several key feature parameters that characterize the material's healing behavior.
[0032] In specific implementation, the process of fitting S2 using a nonlinear model that can characterize the evolution of the healing process from a rapid phase to a slow phase includes: The healing index data measured at different healing times were substituted into the Ramberg-Osgood model equation for nonlinear fitting. By judging the goodness of fit, three characteristic parameters in the equation were determined, which respectively characterize the rapid recovery in the early stage of healing, the slow recovery in the later stage of healing, and the maximum recoverable capacity of the material. These correspond to the short-term healing rate parameter, the long-term healing rate parameter, and the healing potential parameter.
[0033] In specific implementation, in S2, the three characteristic parameters in the equation that characterize rapid recovery in the early stage of healing, slow recovery in the later stage of healing, and the maximum recoverable capacity of the material are determined by judging the goodness of fit. The specific process includes: After substituting the healing index data at different healing times into the Ramberg-Osgood model equation for nonlinear fitting, the model's coefficient of determination is calculated and evaluated by comparing the differences between the calculated values and the experimental values. When the coefficient of determination is higher than the preset accuracy standard, the fitting result is accepted. The parameters in the fitting equation that characterize the healing rate in the early stage of healing, the healing rate in the later stage of healing, and the maximum healing capacity limit are respectively assigned as the short-term healing rate parameter, the long-term healing rate parameter, and the healing potential parameter.
[0034] In this embodiment, the Ramberg-Osgood model equation is: (3) in, Representative process The healing index of the material after the healing time. Represents healing time. Represents the short-term healing rate. Represents the long-term healing rate. Represents healing potential. , and The healing index test results under different healing times were obtained by fitting the formula (3).
[0035] In this embodiment, specifically, the short-term healing rate parameter corresponds to This parameter characterizes the rapid recovery rate in the early stages of healing (at the beginning of healing, the healing rate is high due to crack closure and viscous flow dominating, and this parameter reflects the recovery capacity in this rapid phase).
[0036] Long-term healing rate parameter corresponding This parameter characterizes the slow recovery rate in the later stages of healing (as the healing time increases, the molecular chain diffusion and interface recombination process gradually slows down, and the healing rate tends to stabilize and approach a constant value; this parameter reflects this slow, stable long-term healing rate).
[0037] Healing potential parameters correspond This parameter characterizes the theoretically maximum recoverability limit that a material can achieve (i.e., the asymptotic value that the healing index eventually approaches over time, reflecting the upper limit of the material's self-healing ability under ideal conditions).
[0038] S3. Based on the correspondence between multiple key feature parameters and data of different initial damage states, establish quantitative functional relationships between each feature parameter and the initial damage density to obtain the influence law of damage degree on healing rate and healing potential.
[0039] In specific implementation, the process of establishing the quantitative functional relationship between each characteristic parameter and the initial damage density in S3 includes: For the short-term healing rate parameter, long-term healing rate parameter, and healing potential parameter extracted from S2, power function correlation equations were established between them and the initial damage density. Multiple sets of characteristic parameter data under different initial damage densities were used to fit each power function equation to determine the coefficients in each equation, so as to quantify the decay law of each characteristic parameter as the initial damage density increases.
[0040] In specific implementation, in S3, multiple sets of characteristic parameter data under different initial damage densities are used to fit each power function equation to determine the coefficients in each equation. The specific process includes: For short-term healing rate parameters, long-term healing rate parameters, and healing potential parameters, their corresponding values at multiple different initial damage density levels were collected. With initial damage density as the independent variable and each characteristic parameter as the dependent variable, a nonlinear regression analysis was performed on the data using power function equations. By maximizing the coefficient of determination between the calculated values of the equations and the measured data points, specific power exponent coefficients and proportional coefficients in each power function equation were obtained. In this way, a precise power function decay relationship between each characteristic parameter and the initial damage density was established.
[0041] S4. Substitute the quantitative functional relationship described in S3 into the nonlinear model described in S2, and couple the two variables of initial damage density and healing time to construct a comprehensive mathematical model that can predict the dynamic evolution of the healing index of asphalt material over time under any damage state.
[0042] In specific implementation, in S4, the initial damage density and healing time are coupled to construct a comprehensive mathematical model that can predict the dynamic evolution of the healing index of asphalt materials over time under any damage state. The specific process includes: The three power function relationships obtained in S3, which characterize the short-term healing rate parameter, the long-term healing rate parameter, and the healing potential parameter, respectively, are related to the initial damage density. These are then substituted into the Ramberg-Osgood model equation used in S2 to obtain an explicit prediction equation for the healing index with the initial damage density and healing time as independent variables, which serves as a comprehensive mathematical model.
[0043] In specific implementation, the model application extension step is also included. This step includes: based on the constant reference temperature test data constructed in S4, the comprehensive mathematical model is scaled by introducing a temperature-related shift factor to equivalently scale the healing time variable in the model according to the time-temperature equivalence principle, thereby extending the healing index prediction method to evaluate the healing behavior of asphalt materials under different service temperature conditions.
[0044] S5. The comprehensive mathematical model is used to predict the healing index of asphalt material under the conditions of initial damage state and healing time.
[0045] In practical implementation, the specific process in S5 includes: The initial damage density value and healing time value to be predicted are directly input into the explicit prediction equation corresponding to the comprehensive mathematical model. By substituting the initial damage density value into the calibrated power function relating the characteristic parameters and damage density, the corresponding short-term healing rate parameter value, long-term healing rate parameter value, and healing potential parameter value are calculated. Then, these parameter values, together with the healing time value, are substituted into the Ramberg-Osgood model equation for calculation, and finally, the predicted healing index value under the corresponding conditions is directly output.
[0046] This invention can achieve at least one of the following specific technical effects: (1) Improve the accuracy of fatigue life prediction: Construct a method for predicting the healing performance of asphalt binder that can consider multiple healing time scales and damage level variables. While considering the accumulation of material damage, it can also reflect the partial recovery effect of healing on pavement performance, thereby significantly improving the consistency between predicted life and actual use status, and providing more reliable data for pavement design and life assessment.
[0047] (2) Optimize the timing of pavement maintenance: By quantifying the relationship between healing behavior and damage state and healing time, the timing of maintenance can be made more scientific and precise. This method can avoid premature maintenance and waste of resources caused by ignoring the healing effect, and can also prevent irreversible expansion of structural damage caused by delayed maintenance, thereby improving the efficiency and economy of pavement maintenance management.
[0048] (3) Achieve a quantitative description of the dynamic evolution of healing: Construct a dynamic prediction model of the healing index as it changes with healing time and damage state, so that the model can accurately reflect the time dependence and damage degree dependence of the healing process, and provide a quantitative basis for fatigue life prediction and maintenance strategies.
[0049] (4) Clarify the quantitative relationship between damage and healing: quantify the evolution of key parameters such as short-term healing rate, long-term healing rate and healing potential with the degree of damage, establish a systematic quantitative relationship between the degree of damage and key healing parameters, and be able to characterize the influence of crack closure, interface contact and molecular recombination on healing rate and healing potential parameters under different damage levels.
[0050] (5) A healing index prediction method for engineering practice is provided. In the future, the healing prediction model can be further coupled with the damage evolution model in combination with the actual road surface use conditions to achieve accurate prediction of road fatigue life and optimization of maintenance decisions. This will play an important guiding role in road design, construction and maintenance management and has broad promotion value.
[0051] The specific implementation process of the method of the present invention is as follows: Currently, the healing ability of asphalt materials is usually evaluated using the healing index. The healing index is typically defined based on the relative recovery of material properties before and after healing. The calculation method involves selecting a representative physical quantity (such as complex modulus, storage modulus, fracture energy, or damage density) that characterizes the material's structure or mechanical properties, and quantifying the material's healing recovery ability by calculating the ratio of the recovery amount of this physical quantity during the healing stage to the baseline value at the end of the first fatigue loading. This method can intuitively reflect the relative level of performance recovery during the healing process and is one of the most commonly used indicators for evaluating the healing performance of asphalt binders. A larger healing index (closer to 1) indicates a higher recovery ratio of the material after a certain healing time compared to its initial undamaged state, and a stronger healing ability; conversely, a smaller healing index (closer to 0) indicates a lower recovery ratio of the material after a certain healing time compared to its initial undamaged state, and a weaker healing ability. Since crack damage is essentially a change in the crack length or damage density of the test sample, the change in damage density before and after the healing period is used to calculate the healing index, as shown in formulas (1) and (2).
[0052] (1) (2) in, Representative sample under sinusoidal oscillatory fatigue loading Damage density after one cycle, Representative sample under sinusoidal oscillatory fatigue loading Complex shear modulus after one cycle The complex shear modulus represents the sample at the initial stage of sinusoidal oscillatory fatigue loading. The healing index represents the healing index under given healing conditions. This represents the damage density of a sample under a given sinusoidal oscillatory fatigue load for a certain number of cycles. This represents the initial damage density during the secondary sinusoidal oscillation fatigue loading stage after a certain intermittent healing time.
[0053] After establishing a quantitative calculation method for the healing index based on damage density, it is necessary to further describe its evolution over healing time to reflect the entire process of the material from the rapid recovery stage to the stable stage. The healing process of asphalt binders exhibits significant nonlinear time-varying characteristics: in the early stages of healing, crack closure and viscous flow dominate, resulting in a high healing rate; however, as time progresses, molecular chain diffusion and interfacial reorganization gradually slow down, the healing rate tends to stabilize, and the healing index eventually approaches a stable value. This invention employs the Ramberg-Osgood model to fit the relationship between the healing index and healing time. This model can accurately capture the transition characteristics of the healing process from the rapid to the slow stage and achieve precise characterization of the entire healing evolution process with fewer parameters. By fitting the Osgood model to experimental data under different damage conditions, model parameters such as the short-term healing rate, long-term healing rate, and healing potential of the material at various damage levels can be obtained, laying the foundation for establishing a healing prediction model that considers damage effects.
[0054] Specifically, the Ramberg-Osgood model does not incorporate the influence of injury severity when fitting the relationship between the healing index and healing time. In other words, to fit the healing index-healing time relationship under different injury severity levels using the Ramberg-Osgood model, additional equations are needed to describe the evolution of the three characteristic parameters with varying injury severity.
[0055] (3) in, Representative process The healing index of the material after the healing time. Represents healing time. Represents the short-term healing rate. Represents the long-term healing rate. Represents healing potential. , and The healing index test results under different healing times were obtained by fitting the formula (3).
[0056] Specifically, after analyzing test data from multiple groups with different damage levels, it was found that... , and All three characteristic parameters exhibit a power-law-like decreasing relationship with the change of damage density. Specifically, as the damage density increases, the crack spacing increases, the probability of interfacial contact decreases, and molecular chain recombination is restricted, resulting in a decreasing trend in both the healing rate and potential of the material. Based on this characteristic relationship, this invention establishes quantitative correlation equations between each healing parameter and damage density to characterize the influence of damage on the healing process. Specifically, equations (4) to (6) are used to evaluate the short-term healing rate of asphalt binder under different initial damage densities. Long-term healing rate and healing potential The three parameters are fitted together, and the equation parameters are obtained by solving them separately: (4) (5) (6) Based on this, by substituting the damage dependencies of the above parameters into formula (3), a comprehensive healing index prediction model that simultaneously considers the two core variables of healing time and damage density can be constructed. This model can describe the dynamic evolution of the healing index over time under any damage state, realizing quantitative prediction and parameterized characterization of healing behavior. Through this method, the influence of damage on healing capacity can be revealed, providing a basis for fatigue life correction and optimization of pavement maintenance timing.
[0057] It is the proportional coefficient in formula (4) that characterizes the overall level of short-term healing rate of asphalt binder.
[0058] It is the power exponent in formula (4) that controls how quickly the short-term healing rate decreases as the initial damage density increases.
[0059] It is the proportional coefficient in formula (5) that characterizes the basic value of the long-term healing rate of asphalt adhesive.
[0060] It is the power exponent in formula (5) that reflects the sensitivity of the long-term healing rate to changes in the initial damage density.
[0061] It is the proportional coefficient in formula (6) that characterizes the theoretical maximum value of the healing potential of asphalt binder.
[0062] It is the power exponent in formula (6) that describes the decreasing trend of healing potential as the initial damage density increases.
[0063] Verification Example 1 This verification example uses Grade 70 A base asphalt as the raw material, and the relevant technical indicators of the selected materials all meet the requirements of the "Test Procedures for Asphalt and Asphalt Mixtures in Highway Engineering" (JTG 3410-2025). It is particularly emphasized that the asphalt binder material selected in this verification example is only a representative material to verify the effectiveness of this method, used to illustrate the feasibility of the technical approach and parameter identification process of this invention. In fact, the healing index prediction model and its parameter fitting method established in this invention do not depend on the specific chemical composition of the material, but are based on the universal law of the viscoelastic response, damage evolution, and healing recovery characteristics exhibited by the material during the fatigue loading-healing cycle. Therefore, any polymer material system with similar viscoelastic behavior and capable of reversible damage and self-healing processes can be applied to the analysis and prediction methods proposed in this invention.
[0064] The fatigue-healing-fatigue test is one of the most commonly used experimental methods for evaluating the healing performance of viscoelastic materials. The basic steps are as follows: using a dynamic shear rheometer, a sinusoidal oscillatory cyclic loading is applied to the sample under controlled conditions to induce fatigue damage at the target damage density; then, the sample is left to stand under set healing conditions (such as temperature and healing time) to allow the material to partially recover the structural damage by relying on its own healing properties; finally, the same loading conditions as the first sinusoidal oscillatory cyclic loading are applied again (secondary fatigue loading), and the healing index is quantitatively calculated by comparing the differences in the mechanical response of the material before and after healing according to formula (2) to characterize the healing ability of the material.
[0065] All tests in this verification example were conducted using a dynamic shear rheometer (model: SmartPave 102e). The temperature was maintained at a constant 25°C throughout the test, the strain level under load was set to 4%, and the frequency was set to 10Hz. The specific implementation process of the test was divided into the following three stages: (1) Initial fatigue loading stage Under constant temperature and loading frequency, strain-controlled or stress-controlled cyclic loading is applied to the asphalt binder sample. The initial value of the internal modulus during the first fatigue loading stage is given by formula (1). Then, in the complex shear modulus... Once the damage density decays to the specified value, the program stops loading and enters an intermittent healing phase. This phase of loading is mainly used to generate a controllable damage density within the material, thereby creating healing conditions corresponding to different damage levels.
[0066] (2) Intermittent healing phase After completing step (1) above, the sample is maintained at the set healing temperature and resting time to allow the viscoelastic chain segment movement and interface reconstruction of the material to proceed. After this stage, part of the material's structure can be restored, as evidenced by the recovery of the absolute value of the complex shear modulus.
[0067] (3) Second fatigue loading stage Under the same test conditions as the first fatigue loading stage, sinusoidal oscillatory shear cyclic loading was applied again, and the material response was recorded. Substituting the initial value of the complex shear modulus from the second fatigue loading stage into formula (1) yields the current damage density of the material after healing, where... The complex shear modulus value at the initial stage of the first fatigue loading is still used. Combining formula (2), the healing index result at a specific temperature, damage density, and healing time can be obtained to quantify the self-healing ability of the material.
[0068] To ensure the accuracy of the test results, each sample underwent two parallel tests. The average value was taken as the final test result if the calculated error of the healing index between the two groups under the same conditions was less than 5%. The variables for damage density and healing time during the experiment are shown in Table 1.
[0069] Table 1. Variable settings for the fatigue-healing-fatigue test 1. The Influence of Damage Degree and Healing Time on the Healing Index of Asphalt Binders To explain the experimental basis for the above theoretical method, Figure 1 The test results of the healing index of Grade 70 A base asphalt binder as a function of damage density and healing time are presented. As shown in the figure, when the damage density remains constant, the healing index of the asphalt continuously increases with the increase of healing time (5 min, 10 min, 25 min, 40 min), indicating that the structural integrity and mechanical recovery capacity of the internal system of the material gradually increase during the healing stage. In the early stage of healing, the healing rate is high, mainly due to the rapid movement of molecular chain segments and the reconstruction of interfacial physical contacts; while in the later stage of healing, as the system tends to equilibrium, molecular chain diffusion and interfacial repair gradually slow down, and the increase in the healing index decreases, exhibiting a staged decay characteristic of the healing rate.
[0070] On the other hand, when the healing time is fixed (taking a 40-minute healing time as an example), the healing index of asphalt shows a significant decreasing trend with increasing damage density (0.2, 0.4, 0.6, 0.8). Higher damage levels imply an increase in the number and size of cracks, wider crack spacing, and a reduction in the probability of interfacial contact and effective diffusion channels, thus limiting the recombination and cross-linking ability of molecular chains during the healing stage. Especially under high damage conditions, both the healing rate and healing potential decrease, and the self-healing ability of the material is significantly inhibited. This phenomenon indicates that the healing behavior of asphalt binders exhibits a damage-dependent characteristic; when the damage exceeds a certain threshold, the healing effect of the system is difficult to fully realize.
[0071] In summary, the clear trend of the healing index increasing with healing time and decreasing with damage severity clearly verifies the rationality of the modeling approach in this invention, namely, that healing behavior is simultaneously influenced by the coupled effects of initial damage density and healing time. This result provides an experimental basis and data support for subsequently establishing quantitative models of healing rate, long-term healing potential, and their relationship with damage density.
[0072] 2. A method for predicting the healing index that couples damage severity with healing time. Based on the model and experimental test results, the healing index test results under different initial damage densities and healing times were first fitted using formula (3) to obtain the short-term healing rate, long-term healing rate, and healing potential parameters, as shown in Table 2. Then, formulas (4) to (6) were used to fit the healing characteristic parameters under different initial damage densities, as shown in Table 3. The R-squared values of the fitted equations... 2 All values are greater than 0.963, indicating that the power functions of formulas (4) to (6) can reasonably describe the evolution of the three healing characteristic parameters with the initial damage density. Subsequently, by substituting the obtained healing characteristic parameter equations into formula (3), a healing index prediction model that comprehensively considers the influence of damage density and healing time can be obtained, as shown in formula (7). The physical meaning of each parameter in formula (7) is the same as that described in the theoretical method section above.
[0073] Table 2 Fitting results of healing characteristic parameters Table 3. Parameter Fitting Results of Healing Characteristic Parameter Equations (7) To verify the effectiveness of the theoretical prediction model of formula (7), Figure 2 The figure shows the comparison between the calculated healing index obtained according to formula (7) and the experimental test values. It can be seen from the figure that the model calculated values and the measured values show good consistency under different damage degrees and healing time conditions. The data points are generally distributed near the fitted line, indicating that the constructed model can accurately reflect the healing law of the material. The coefficient of determination R between the calculated and measured values is shown. 2 A value greater than 0.9929 indicates that the healing index prediction model for the coupling damage degree and healing time proposed in this invention has good fitting accuracy and reliability.
[0074] Furthermore, this model maintains high stability and universality under different healing times and damage levels, providing a reference for the healing behavior analysis of other polymer materials with similar viscoelastic properties. Therefore, the healing index model proposed in this invention not only enables high-precision characterization of the healing behavior of asphalt binders under different damage degrees and healing times, but also provides reliable theoretical support for fatigue life prediction, structural design optimization, and performance evaluation of recycled materials. The model exhibits good physical interpretability and computational simplicity in its parameter form, and can be extended to other polymer material systems with viscoelastic, damage, and healing characteristics, providing a new analytical approach and model foundation for the study of healing mechanisms and engineering applications of such materials.
[0075] The above description of the embodiments is provided to enable those skilled in the art to understand and use the invention. It will be apparent to those skilled in the art that various modifications can be made to these embodiments, and the general principles described herein can be applied to other embodiments without inventive effort. Therefore, the present invention is not limited to the above embodiments, and any improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope of the invention should be within the protection scope of the present invention.
Claims
1. A method for predicting the healing performance of asphalt materials based on the damage-time coupling effect, characterized in that, Includes the following steps: S1. Through cyclic tests of fatigue loading, intermittent healing, and re-fatigue loading, the initial damage density and healing index of asphalt materials under different initial damage states and different healing times were obtained. S2. Based on the obtained data on the corresponding healing index and healing time, a nonlinear model that can characterize the evolution of the healing process from a rapid stage to a slow stage is used for fitting, thereby extracting multiple key feature parameters that characterize the material's healing behavior. S3. Based on the correspondence between multiple key feature parameters and data of different initial damage states, establish quantitative functional relationships between each feature parameter and the initial damage density to obtain the influence law of damage degree on healing rate and healing potential. S4. Substitute the quantitative functional relationship described in S3 into the nonlinear model described in S2, and couple the two variables of initial damage density and healing time to construct a comprehensive mathematical model that can predict the dynamic evolution of the healing index of asphalt material over time under any damage state. S5. The comprehensive mathematical model is used to predict the healing index of asphalt material under the conditions of initial damage state and healing time.
2. The method for predicting the healing performance of asphalt materials based on the damage-time coupling effect according to claim 1, characterized in that, In S1, the specific process for obtaining the initial damage density and healing index of asphalt materials under different initial damage states and different healing times includes: The asphalt material sample was subjected to the first sinusoidal oscillatory fatigue loading until the complex shear modulus of the material decayed to the corresponding preset damage density and the loading was stopped. The sample was left to stand at a constant temperature for several pre-set different durations to achieve intermittent healing; A sinusoidal oscillatory fatigue load with the same conditions as the first loading was applied again, and the initial damage density and healing index were calculated based on the initial complex shear modulus of the first loading, the complex shear modulus at the end of the first loading, and the initial complex shear modulus of the second loading.
3. The method for predicting the healing performance of asphalt materials based on the damage-time coupling effect according to claim 2, characterized in that, In S1, the damage density and healing index are calculated based on the initial complex shear modulus after the first loading, the complex shear modulus at the end of the first loading, and the initial complex shear modulus after the second loading. The specific process includes: The damage density at the end of the first fatigue is calculated based on the ratio of the complex shear modulus at the end of the first loading to the initial complex shear modulus at the beginning of the first loading. Then, the damage density at the end of intermittent healing is calculated based on the ratio of the initial complex shear modulus at the beginning of the second loading to the initial complex shear modulus at the beginning of the first loading. The damage density at the end of the intermittent healing is compared with the damage density at the end of the first fatigue. The healing index, which characterizes the degree of recovery of the material's self-healing ability under the corresponding healing conditions, is obtained by calculating the ratio of the difference between the two to the damage density at the end of the first fatigue.
4. The method for predicting the healing performance of asphalt materials based on the damage-time coupling effect according to claim 1, characterized in that, In S2, the specific process of fitting the model using a nonlinear model that can characterize the evolution of the healing process from a rapid phase to a slow phase includes: The healing index data measured at different healing times were substituted into the Ramberg-Osgood model equation for nonlinear fitting. By judging the goodness of fit, three characteristic parameters in the equation were determined, which respectively characterize the rapid recovery in the early stage of healing, the slow recovery in the later stage of healing, and the maximum recoverable capacity of the material. These correspond to the short-term healing rate parameter, the long-term healing rate parameter, and the healing potential parameter.
5. The method for predicting the healing performance of asphalt materials based on the damage-time coupling effect according to claim 4, characterized in that, In S2, by determining the goodness of fit, three characteristic parameters in the equation are identified, representing rapid recovery in the early stage of healing, slow recovery in the later stage of healing, and the maximum recoverable capacity of the material, respectively. The specific process includes: After substituting the healing index data at different healing times into the Ramberg-Osgood model equation for nonlinear fitting, the model's coefficient of determination is calculated and evaluated by comparing the differences between the calculated values and the experimental values. When the coefficient of determination is higher than the preset accuracy standard, the fitting result is accepted. The parameters in the fitting equation that characterize the healing rate in the early stage of healing, the healing rate in the later stage of healing, and the maximum healing capacity limit are respectively assigned as the short-term healing rate parameter, the long-term healing rate parameter, and the healing potential parameter.
6. The method for predicting the healing performance of asphalt materials based on the damage-time coupling effect according to claim 4, characterized in that, In S3, the specific process of establishing the quantitative functional relationship between each characteristic parameter and the initial damage density includes: For the short-term healing rate parameter, long-term healing rate parameter, and healing potential parameter extracted from S2, power function correlation equations were established between them and the initial damage density. Multiple sets of characteristic parameter data under different initial damage densities were used to fit each power function equation to determine the coefficients in each equation, so as to quantify the decay law of each characteristic parameter as the initial damage density increases.
7. The method for predicting the healing performance of asphalt materials based on the damage-time coupling effect according to claim 6, characterized in that, In S3, multiple sets of characteristic parameter data under different initial damage densities are used to fit each power function equation to determine the coefficients in each equation. The specific process includes: For short-term healing rate parameters, long-term healing rate parameters, and healing potential parameters, their corresponding values at multiple different initial damage density levels were collected. With initial damage density as the independent variable and each characteristic parameter as the dependent variable, a nonlinear regression analysis was performed on the data using power function equations. By maximizing the coefficient of determination between the calculated values of the equations and the measured data points, specific power exponent coefficients and proportional coefficients in each power function equation were obtained, thereby establishing a power function decay relationship between each characteristic parameter and the initial damage density.
8. The method for predicting the healing performance of asphalt materials based on the damage-time coupling effect according to claim 7, characterized in that, In S4, by coupling the initial damage density and healing time, a comprehensive mathematical model is constructed that can predict the dynamic evolution of the healing index of asphalt materials over time under any damage state. The specific process includes: The three power function relationships representing the short-term healing rate parameter, long-term healing rate parameter, and healing potential parameter related to the initial damage density are substituted into the Ramberg-Osgood model equation used in S2 to obtain an explicit prediction equation for the healing index with the initial damage density and healing time as independent variables, which serves as a comprehensive mathematical model.
9. The method for predicting the healing performance of asphalt materials based on the damage-time coupling effect according to claim 1, characterized in that, It also includes a model application extension step, which includes: based on the constant reference temperature test data constructed in S4, the comprehensive mathematical model is scaled by introducing a temperature-related shift factor to equivalently scale the healing time variable in the model according to the time-temperature equivalence principle, thereby extending the healing index prediction method to evaluate the healing behavior of asphalt materials under different service temperature conditions.
10. The method for predicting the healing performance of asphalt materials based on the damage-time coupling effect according to claim 7, characterized in that, In S5, the specific process includes: The initial damage density value and healing time value to be predicted are directly input into the explicit prediction equation corresponding to the comprehensive mathematical model. By substituting the initial damage density value into the calibrated power function relating the characteristic parameters and damage density, the corresponding short-term healing rate parameter value, long-term healing rate parameter value, and healing potential parameter value are calculated. Then, these parameter values, together with the healing time value, are substituted into the Ramberg-Osgood model equation for calculation, and finally, the predicted healing index value under the corresponding conditions is directly output.