A Method for Assessing Prestress Loss in Prestressed Beams Based on Digital Simulation

Abstract

This invention provides a method for assessing prestress loss in prestressed beams based on digital simulation, relating to the field of concrete structure performance assessment technology. The specific steps include: first, establishing a finite element model and dividing the beam into several sub-regions, calculating the regional sensitivity coefficient by combining the material properties and environmental parameters of each region; second, collecting measured data to invert and calibrate the model, optimizing duct friction and anchor deformation parameters; subsequently, inputting historical environmental and load sequences, calculating the time-varying loss of each sub-region through time-step simulation, and summarizing to generate a comprehensive loss index for the entire region; further, quantifying the loss components caused by concrete shrinkage, creep, and prestressing tendon relaxation through independent simulation; and finally, predicting the loss development trend based on future working conditions, providing a reliable basis for long-term structural performance assessment and maintenance decisions.

Description

Technical Field

[0001] This invention relates to the field of concrete structure performance evaluation technology, specifically a method for evaluating prestress loss in prestressed beams based on digital simulation. Background Technology

[0002] Prestressed beams are widely used in modern bridges and building structures. The loss of prestress during their service is a key factor affecting the durability and safety of the structure. The loss of prestress is mainly caused by factors such as concrete shrinkage, creep, and relaxation of prestressing tendons. These time-varying effects will cause the effective prestress of the beam to gradually decrease, thereby affecting the long-term service performance of the structure. Traditional methods for assessing prestress loss mostly rely on empirical formulas in specifications or local on-site monitoring, which are difficult to fully reflect the spatiotemporal variation characteristics of the structure under actual environment and load. Especially when facing complex environments, material non-homogeneity, and long-term performance prediction, they have problems such as insufficient accuracy and poor adaptability.

[0003] In the prior art, the deep learning-based method for predicting long-term prestress loss during staged tensioning, disclosed in CN120874462A, achieves the prediction of prestress loss through finite element simulation combined with multilayer perceptron neural network. Although this method improves the automation of prediction to a certain extent, it relies on a large amount of finite data of preset working conditions and lacks a detailed characterization of regional differences, time-varying material properties, and dynamic influence of environmental parameters in actual structures. In addition, the model calibration process of this method does not fully consider the actual influence of construction process parameters such as duct friction and anchor deformation, which limits its applicability and accuracy in real structures and makes it difficult to achieve independent analysis and trend prediction of prestress loss components (shrinkage, creep, and prestressing tendon relaxation).

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for evaluating the prestress loss of prestressed beams based on digital simulation, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for assessing prestress loss in prestressed beams based on digital simulation, comprising the following steps:

[0008] Step 1: Construct a finite element model of the prestressed beam and divide the finite element model into multiple sub-regions along the length of the beam. For each sub-region, determine the region sensitivity coefficient to characterize the time-varying properties of the material in that sub-region based on its material parameters and historical environmental parameters.

[0009] Step 2: Collect the initial strain and pressure data of each sub-region at the starting point of the time step calculation, compare them with the initial simulation data of the model, and adjust the duct friction coefficient and anchor deformation value in the model through inversion analysis to complete the model calibration;

[0010] Step 3: Input the historical environmental parameters and historical load data of each sub-region from the time step calculation start point to the current time into the calibrated model in time series, drive the model to perform time step calculation, simulate the structural response of each sub-region under the corresponding historical environment and load, calculate the time-varying loss value of each sub-region based on the structural response and combined with the regional sensitivity coefficient, and summarize the time-varying loss values ​​of all sub-regions to generate the global comprehensive loss index.

[0011] Step 4: For each sub-region, perform independent derivation calculations considering only concrete shrinkage, only creep, and only prestressing tendon relaxation. By summarizing the independent derivation results of each sub-region, obtain the first loss component, second loss component, and third loss component of the beam corresponding to shrinkage, creep, and prestressing tendon relaxation, respectively.

[0012] Step 5: Based on the preset future environmental parameters and expected loads, drive the calibrated model to perform forward extrapolation and predict the changing trends of the global comprehensive loss index, the first loss component, the second loss component, and the third loss component at the future target time point.

[0013] Furthermore, the specific execution process of step 1 is as follows:

[0014] Based on the design and construction drawings of the prestressed beam, the geometric dimensions of the beam, the design strength grade of the concrete, the modulus of elasticity, the arrangement of the prestressed steel strands and the material properties are obtained. A parametric finite element model is established in the finite element analysis software. Then, the finite element model is evenly divided into multiple sub-regions along the length of the beam, and the division spacing is determined according to the length of the beam.

[0015] For each sub-region after division, its material parameters and historical environmental parameters are collected. The material parameters include the actual compressive strength and volumetric reinforcement ratio of the concrete in the sub-region. The actual compressive strength is obtained by ultrasonic rebound combined method, and the volumetric reinforcement ratio is determined based on design drawings and construction acceptance records. The historical environmental parameters refer to the daily average temperature and daily average relative humidity of the sub-region in the past month, which are obtained by historical monitoring records of temperature and humidity sensors deployed at corresponding locations on the beam.

[0016] For each sub-region, based on its material parameters and historical environmental parameters, a regional sensitivity coefficient is determined to characterize the time-varying properties of the material in that sub-region. The calculation formula is as follows:

[0017]

[0018] In the formula, For the first Regional sensitivity coefficient of each sub-region For indexes of sub-regions; For the first The actual compressive strength of concrete in each sub-region, expressed in megapascals (MPa). This is a reference value for the compressive strength of concrete, expressed in megapascals (MPa). For the first Volumetric reinforcement ratio of each sub-region; The first time in the past month The standard deviation of the daily average temperature for each sub-region, in degrees Celsius; The first time in the past month The average daily temperature of each sub-region, in degrees Celsius; The first time in the past month The standard deviation of the daily average relative humidity for each sub-region The first time in the past month The average daily relative humidity of each sub-region; , and The preset weights for the corresponding indicators, And satisfy .

[0019] Furthermore, a time period is defined, with the current time as the starting point. As the endpoint, a one-month time period is traced back, and the starting point of this period is marked as the starting point of the time step calculation. At the starting point of the time step calculation, for each sub-region, pressure sensors are arranged at the corresponding prestressed steel strand anchorage end to collect the anchorage pressure data. At the same time, strain gauges are arranged at the mid-span section, quarter-point section, and support section of each sub-region to collect concrete surface strain data. The initial strain and pressure data of each sub-region at the starting point of the time step calculation are used to form a measured dataset, and the corresponding data calculated by the finite element model under the same boundary conditions are used to form an initial simulation dataset.

[0020] The measured dataset is compared with the initial simulated dataset item by item, and the relative error of each data point is calculated. The objective function of the inversion analysis is established with the goal of minimizing the sum of squares of the relative errors of all data points. The duct friction coefficient and anchor deformation value are used as variables to be optimized, and the gradient descent method is used for iterative optimization.

[0021] In each iteration, the parameters of the finite element model are updated according to the current optimization variables, the initial simulation dataset is recalculated and the objective function value is evaluated. When the objective function value is less than the preset threshold, the optimization stops. The obtained duct friction coefficient and anchor deformation value are the optimal solutions, thus completing the calibration of the finite element model.

[0022] Furthermore, the data collected from each sub-region from the start point of time step calculation to the current time will be... Historical environmental parameters and historical load data are arranged in chronological order to form a complete time series, which is then input into the calibrated finite element model. The historical environmental parameters refer to the data from the start of the time step calculation to the current time in the sub-region. The daily average temperature and daily average relative humidity during this period; the historical load data are the maximum daily equivalent uniformly distributed loads acting on the beam during the corresponding period, collected by the structural health monitoring system.

[0023] The finite element model is driven to perform time-step calculations according to the input time series, simulating the structural response of each sub-region at each time step. The structural response includes the stress time history and strain time history of the concrete. Based on the simulated structural response and the region sensitivity coefficient calculated in step 1, the time-varying loss value of each sub-region is calculated using the following formula:

[0024]

[0025] In the formula, For the first Each sub-region at the current moment The time-varying loss value; For the first Regional sensitivity coefficient of each sub-region; For the first Each sub-region at time The concrete stress was obtained through finite element simulation; For the first Each sub-region at time The concrete creep rate was calculated based on the CEB-FIP model specification. For the first The elastic modulus of concrete in each sub-region was obtained through laboratory material testing. For the first Each sub-region at time The concrete shrinkage rate was calculated based on the CEB-FIP model specification. This serves as the starting point for time step calculation; For integration time;

[0026] All sub-regions at the same time The time-varying loss values ​​are weighted and summed to obtain the global comprehensive loss index, based on the following formula:

[0027]

[0028] In the formula, For the current moment The overall loss index of the entire domain; For the first The weighting coefficients of each sub-region are determined based on the volume ratio of each sub-region. This represents the total number of sub-regions.

[0029] Furthermore, three independent calculation modules were established in the calibrated finite element model: a concrete shrinkage calculation module, used to calculate concrete shrinkage strain according to the CEB-FIP model specification; a concrete creep calculation module, used to calculate concrete creep strain according to the CEB-FIP model specification; and a prestressed tendon relaxation calculation module, used to calculate stress relaxation loss based on the prestressed tendon material properties.

[0030] For each sub-region, three independent simulation calculations are performed: the first simulation activates only the concrete shrinkage calculation module, while keeping the creep and prestressed tendon relaxation calculation modules in an inactive state; the second simulation activates only the concrete creep calculation module, while keeping the shrinkage and prestressed tendon relaxation calculation modules in an inactive state; the third simulation activates only the prestressed tendon relaxation calculation module, while keeping the shrinkage and creep calculation modules in an inactive state.

[0031] Each independent simulation uses the same historical environmental parameters and historical load data as in step 3 as input, and calculates the loss components of each sub-region using the following formula:

[0032] For the first simulation:

[0033]

[0034] In the formula, Indicates the first Each sub-region at the current moment The loss component caused by concrete shrinkage;

[0035] For the second deduction:

[0036]

[0037] In the formula, Indicates the first Each sub-region at the current moment The loss component caused by creep;

[0038] For the third simulation:

[0039]

[0040] In the formula, Indicates the first Each sub-region at the current moment The loss component caused by prestressing tendon relaxation; The stress relaxation amount of the prestressing tendon is calculated based on the prestressing tendon material relaxation test data.

[0041] The individual loss components of each sub-region are weighted and summed to obtain the first, second, and third loss components corresponding to the beam. The formulas used are as follows:

[0042]

[0043] In the formula, Indicates the current moment The first loss component;

[0044]

[0045] In the formula, Indicates the current moment The second loss component;

[0046]

[0047] In the formula, Indicates the current moment The third loss component.

[0048] Furthermore, based on the calibrated finite element model that has completed historical extrapolation and loss component calculation, the preset future environmental parameters and expected loads are used as input conditions; the future environmental parameters include the daily average temperature and daily average relative humidity of each day in the future target time period obtained based on meteorological forecast data; the expected loads include the maximum value of the equivalent uniformly distributed load of each day in the future target time period determined based on design usage requirements.

[0049] The calibrated model is then used for forward extrapolation calculations to predict the global comprehensive loss index at future target time points using the following formula:

[0050]

[0051] In the formula, For future target time points The predicted value of the overall loss index across the entire region; For the current moment The overall loss index of the entire domain; Indicates from the current moment To the future target time point The incremental loss within the period is calculated using the following formula:

[0052]

[0053] in, For concrete stress predicted based on expected load; The creep rate of concrete is predicted based on future environmental parameters; This refers to the concrete shrinkage rate predicted based on future environmental parameters.

[0054] Furthermore, based on the loss component calculation method in step 4, the changing trends of the first, second, and third loss components at future target time points are predicted respectively. The specific logic is as follows:

[0055] First loss component prediction:

[0056]

[0057] In the formula, For future target time points The predicted value of the first loss component;

[0058] Second loss component prediction:

[0059]

[0060] In the formula, For future target time points The predicted value of the second loss component;

[0061] Third loss component prediction:

[0062]

[0063] In the formula, For future target time points The predicted value of the third loss component;

[0064] Based on the calculated future target time points The predicted value of the overall loss index First loss component prediction value Predicted value of the second loss component and the predicted value of the third loss component The changing trends of each parameter can be obtained in the following ways:

[0065] The current moment's overall loss index With future target time points The predicted value of the overall loss index By comparing and calculating their absolute changes, the growth rate of the overall loss index can be characterized.

[0066] The first, second, and third loss components at the current moment are compared with the future target time point. The predicted values ​​of the first, second, and third loss components are compared to calculate their absolute changes. By comparing the magnitude of the absolute changes of each loss component, the dominant factor contributing the most to the growth of total loss is identified. By setting multiple consecutive future target time points for repeated prediction, the overall loss index and the sequence data of each loss component changing over time are obtained. Based on this sequence data, the correspondence between time and loss value is established, and the trend function of each parameter is obtained by curve fitting, thereby quantifying its long-term development law.

[0067] Compared with the prior art, the beneficial effects of the present invention are:

[0068] First, by dividing the prestressed beam into multiple sub-regions along its length and introducing a regional sensitivity coefficient, this invention achieves the characterization of the spatial variability and time-varying characteristics of the structure. This method can more accurately reflect the response differences of different sub-regions under real environment and load, overcome the prediction bias caused by the traditional method of treating the beam as a uniform body, and significantly improve the spatial resolution and accuracy of prestress loss assessment.

[0069] Secondly, this invention proposes a finite element model inversion calibration mechanism based on measured data. By optimizing key construction parameters such as the duct friction coefficient and anchor deformation value, the digital model is made to be highly consistent with the actual structural behavior. Compared with existing methods that rely on preset working conditions or pure theoretical simulation, this method greatly improves the applicability and prediction credibility of the model in actual engineering, and lays a reliable foundation for subsequent long-term performance extrapolation.

[0070] Furthermore, this invention constructs an evaluation framework that combines independent deduction of individual components with comprehensive prediction of the entire domain. It can quantify the loss components caused by shrinkage, creep, and relaxation of prestressing tendons, and predict future trends based on a calibration model. This framework not only enables accurate tracing of the causes of prestress loss and identification of dominant factors, but also provides a scientific basis for decision-making on targeted maintenance and tension compensation strategies for structures. Attached Figure Description

[0071] Figure 1 This is a schematic diagram of the overall method flow of the present invention;

[0072] Figure 2 3D scatter plots of actual compressive strength, volumetric reinforcement ratio, and regional sensitivity coefficient;

[0073] Figure 3This is a parallel coordinate graph of the standard deviation of daily average temperature, the standard deviation of daily average relative humidity, and the regional sensitivity coefficient. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0075] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0076] Example:

[0077] Please see Figures 1-3 The present invention provides a technical solution:

[0078] A method for assessing prestress loss in prestressed beams based on digital simulation, comprising the following steps:

[0079] Step 1: Construct a finite element model of the prestressed beam and divide the finite element model into multiple sub-regions along the length of the beam. For each sub-region, determine the region sensitivity coefficient to characterize the time-varying properties of the material in that sub-region based on its material parameters and historical environmental parameters.

[0080] In this embodiment, the specific execution process of step 1 is as follows:

[0081] Based on the design and construction drawings of the prestressed beam, the geometric dimensions of the beam, the design strength grade of the concrete, the modulus of elasticity, the arrangement of the prestressed steel strands and the material properties are obtained. A parametric finite element model is established in the finite element analysis software. Then, the finite element model is evenly divided into multiple sub-regions along the length of the beam, and the division spacing is determined according to the length of the beam.

[0082] For each sub-region after division, its material parameters and historical environmental parameters are collected. The material parameters include the actual compressive strength and volumetric reinforcement ratio of the concrete in the sub-region. The actual compressive strength is obtained by ultrasonic rebound combined method, and the volumetric reinforcement ratio is determined based on design drawings and construction acceptance records. The historical environmental parameters refer to the daily average temperature and daily average relative humidity of the sub-region in the past month, which are obtained by historical monitoring records of temperature and humidity sensors deployed at corresponding locations on the beam.

[0083] For each sub-region, based on its material parameters and historical environmental parameters, a regional sensitivity coefficient is determined to characterize the time-varying properties of the material in that sub-region. The calculation formula is as follows:

[0084]

[0085] In the formula, For the first Regional sensitivity coefficient of each sub-region For indexes of sub-regions; For the first The actual compressive strength of concrete in each sub-region, expressed in megapascals (MPa). This is a reference value for the compressive strength of concrete, expressed in megapascals (MPa). For the first Volumetric reinforcement ratio of each sub-region; The first time in the past month The standard deviation of the daily average temperature for each sub-region, in degrees Celsius; The first time in the past month The average daily temperature of each sub-region, in degrees Celsius; The first time in the past month The standard deviation of the daily average relative humidity for each sub-region The first time in the past month The average daily relative humidity of each sub-region; , and The preset weights for the corresponding indicators, And satisfy .

[0086] In prestressed beams, the actual compressive strength of concrete is the most important factor determining the instantaneous load-bearing capacity and long-term creep and shrinkage characteristics of the structure. It typically contributes the most to prestress loss and is therefore given a weight. The volumetric reinforcement ratio is the most important factor; it reflects the restraining effect of steel bars on concrete deformation and directly affects stress distribution and creep development. Its effect is secondary, hence its lower weight. The weighting is moderate; although temperature and humidity fluctuations affect the shrinkage and creep rates of concrete, the magnitude of these fluctuations has a relatively weaker and slower impact on losses, therefore the weighting is moderate. Minimum.

[0087] For this formula, the dependent variable Used to characterize the The sensitivity of materials in individual sub-regions to prestress loss under time-varying environments and loads. A higher value indicates that the region is more sensitive to time-varying effects such as concrete shrinkage, creep, and prestressing tendon relaxation, and may experience greater prestress loss under the same environmental and load conditions; conversely, a lower value indicates a lower prestress loss. The smaller the value, the more stable the material properties in that region are and the lower the risk of time-varying loss.

[0088] When the actual compressive strength of concrete At lower levels, its microstructure is relatively loose, with greater potential for contraction and creep, thus leading to a higher regional sensitivity coefficient. The increase reflects the enhanced sensitivity of the material to time-varying losses in this region; the volumetric reinforcement ratio While increasing the reinforcement can enhance local stiffness, it also strengthens the restraining effect of the steel bars on concrete deformation, generating greater internal stress during shrinkage and creep, which can also lead to… The increase indicates that the constraint effect exacerbates the risk of prestress loss; while the temperature and humidity fluctuation term... An increase in this value indicates unstable environmental conditions, which exacerbate the migration of moisture within the concrete and the effects of thermal expansion and contraction, thereby promoting the development of shrinkage and creep. Therefore, an increase in this value will directly lead to… An increase in the value indicates that the region is more sensitive to changes in the external environment and is more prone to time-varying losses.

[0089] The formula uses a linear weighted combination form. , , The three components respectively characterize material strength, reinforcement constraints, and environmental fluctuations, covering the main time-varying factors affecting prestress loss; through weighting... , and Distinguishing the contribution of each factor ensures that the model has both theoretical basis and adjustability.

[0090] Table 1: Statistics of Regional Sensitivity Coefficients

[0091]

[0092] Based on the analysis of the 15 sets of data provided in Table 1 above, it can be seen that the constructed regional sensitivity coefficient model can systematically integrate and characterize the key regional differences that affect prestress loss. The data analysis clearly reveals the inherent correlation between the regional sensitivity coefficient and each input parameter: in sub-regions with lower actual compressive strength of concrete, higher volumetric reinforcement ratio, and more drastic temperature and humidity fluctuations in the service environment, the calculated sensitivity coefficient increases significantly. This quantitative relationship accurately reflects the comprehensive influence of different material properties and external environment on the sensitivity of concrete to time-varying effects such as shrinkage and creep, which is consistent with the structural time-varying theory.

[0093] Furthermore, the analysis confirms the effectiveness of the present invention in achieving refined evaluation by introducing sensitivity coefficients. The data shows that there are considerable differences in sensitivity coefficients between different sub-regions, which directly reflects the non-uniformity of the potential risk of prestress loss in the spatial distribution of the beam. Integrating this differentiated coefficient into the subsequent finite element digital simulation enables the method to surpass the traditional homogenization model and achieve a more realistic and spatially resolved simulation and prediction of the evolution of loss along the beam length, thereby providing a reliable quantitative basis for the accurate diagnosis of structural performance and maintenance decisions.

[0094] Step 2: Collect the initial strain and pressure data of each sub-region at the starting point of the time step calculation, compare them with the initial simulation data of the model, and adjust the duct friction coefficient and anchor deformation value in the model through inversion analysis to complete the model calibration;

[0095] In this embodiment, a time period is defined, with the current time as the starting point. As the endpoint, a one-month time period is traced back, and the starting point of this period is marked as the starting point of the time step calculation. At the starting point of the time step calculation, for each sub-region, pressure sensors are arranged at the corresponding prestressed steel strand anchorage end to collect the anchorage pressure data. At the same time, strain gauges are arranged at the mid-span section, quarter-point section, and support section of each sub-region to collect concrete surface strain data. The initial strain and pressure data of each sub-region at the starting point of the time step calculation are used to form a measured dataset, and the corresponding data calculated by the finite element model under the same boundary conditions are used to form an initial simulation dataset.

[0096] The measured dataset was compared item by item with the initial simulated dataset, and the relative error of each data point was calculated. The objective function for the inversion analysis was established with the goal of minimizing the sum of squared relative errors of all data points. The duct friction coefficient and anchor deformation were used as the variables to be optimized, and the gradient descent method was employed for iterative optimization, with an iteration step size set to [value missing]. The initial values ​​of the duct friction coefficient and the anchor deformation value are set based on the design plan and engineering experience.

[0097] In each iteration, the finite element model parameters are updated according to the current optimization variables, the initial simulation dataset is recalculated and the objective function value is evaluated. When the objective function value is less than the preset threshold, the optimization stops. The obtained duct friction coefficient and anchor deformation value at this time are the optimal solutions, thus completing the calibration of the finite element model. This calibration process lays a reliable foundation for subsequent time-stepping simulations.

[0098] Step 3: Input the historical environmental parameters and historical load data of each sub-region from the time step calculation start point to the current time into the calibrated model in time series, drive the model to perform time step calculation, simulate the structural response of each sub-region under the corresponding historical environment and load, calculate the time-varying loss value of each sub-region based on the structural response and combined with the regional sensitivity coefficient, and summarize the time-varying loss values ​​of all sub-regions to generate the global comprehensive loss index.

[0099] In this embodiment, the data collected from the time step calculation start point to the current time in each sub-region will be used. Historical environmental parameters and historical load data are arranged in chronological order to form a complete time series, which is then input into the calibrated finite element model. The historical environmental parameters refer to the data from the start of the time step calculation to the current time in the sub-region. The daily average temperature and daily average relative humidity during this period; the historical load data are the maximum daily equivalent uniformly distributed loads acting on the beam during the corresponding period, collected by the structural health monitoring system.

[0100] The finite element model is driven to perform time-step calculations according to the input time series, simulating the structural response of each sub-region at each time step. The structural response includes the stress time history and strain time history of the concrete. Based on the simulated structural response and the region sensitivity coefficient calculated in step 1, the time-varying loss value of each sub-region is calculated using the following formula:

[0101]

[0102] In the formula, For the first Each sub-region at the current moment The time-varying loss value; For the first Regional sensitivity coefficient of each sub-region; For the first Each sub-region at time The concrete stress was obtained through finite element simulation; For the first Each sub-region at time The concrete creep rate was calculated based on the CEB-FIP model specification. For the first The elastic modulus of concrete in each sub-region was obtained through laboratory material testing. For the first Each sub-region at time The concrete shrinkage rate was calculated based on the CEB-FIP model specification. This serves as the starting point for time step calculation; This is the time variable for integration.

[0103] For this formula, the dependent variable Used to characterize the Each sub-region at the current moment The time-varying prestress loss caused by the combined effects of concrete shrinkage and creep is a quantitative indicator of the accumulated loss value in this area under historical environmental and load conditions. The larger the value, the more severe the prestress loss that has occurred in the sub-region at the current moment, and the higher the risk of local structural performance degradation; conversely, the smaller the value, the better the prestress is maintained in the region, and the less the time-varying effect is.

[0104] When the regional sensitivity coefficient An increase in this value indicates that the material in that sub-region is more sensitive to time-varying effects, which will amplify the cumulative losses caused by concrete shrinkage and creep, thus directly affecting the dependent variable. Increase; concrete stress An increase in will enhance the creep driving force, thus accelerating loss development at the same creep rate; therefore, the dependent variable Increases with increasing stress; concrete creep rate The acceleration of creep means an increase in irreversible deformation per unit time, which directly exacerbates the accumulation of prestress loss caused by creep and promotes the dependent variable Growth; and This reflects the rate of elastic stress loss caused by shrinkage. The faster the shrinkage rate or the greater the elastic modulus of concrete, the more significant the prestress loss due to shrinkage, which in turn drives the dependent variable. Increase.

[0105] This formula uses an integral form to accumulate the loss rate over the time history, which is consistent with the physical nature of the cumulative effect of time-varying effects; the formula clearly distinguishes creep loss. and shrinkage loss Two key components cover the core causes of prestress loss; a regional sensitivity coefficient is introduced. As a pre-weight, it reflects the modulating effect of material and environmental differences in different regions on loss development, enhancing the model's regional resolution and engineering applicability; the overall structure is clear and the physical meaning is explicit, making it easy to combine with finite element simulation and standard models, and it has good theoretical rationality and practicality.

[0106] All sub-regions at the same time The time-varying loss values ​​are weighted and summed to obtain the global comprehensive loss index, based on the following formula:

[0107]

[0108] In the formula, For the current moment The overall loss index of the entire domain; For the first The weight coefficient of each sub-region is determined based on the volume ratio of each sub-region. Specifically, the proportion of the volume of each sub-region to the total volume of the beam is calculated, and this proportion is used as the weight coefficient of the corresponding sub-region. It is ensured that the sum of the weight coefficients of all sub-regions is 1. This represents the total number of sub-regions.

[0109] Used to represent the current time The overall prestress loss of the prestressed beam is a global index obtained by weighted summation of the time-varying loss values ​​of each sub-region. A higher value indicates a more severe loss of prestress in the beam, a higher risk of overall structural performance degradation, and may affect its load-bearing capacity and long-term durability; conversely, a lower value indicates a lower prestress loss. The smaller the value, the better the overall prestress of the beam is maintained, the smaller the overall loss caused by time-varying effects, and the more safe the structure is.

[0110] As a comprehensive loss index covering the entire region, it directly depends on the time-varying loss values ​​of each sub-region. When a certain sub-region An increase in this value indicates a greater prestress loss in that region due to time-varying effects such as concrete shrinkage and creep. This is typically caused by poorer material properties, higher stress levels, or more severe environmental influences in that region. During the weighted summation process, the increase in loss in this sub-region directly contributes to... In the middle, the time-varying loss index increases, reflecting the cumulative impact of beam performance degradation on overall safety in this region; conversely, if the time-varying loss values ​​of each sub-region are relatively small, then... The corresponding reduction indicates that the overall loss is well controlled.

[0111] Step 4: For each sub-region, perform independent derivation calculations considering only concrete shrinkage, only creep, and only prestressing tendon relaxation. By summarizing the independent derivation results of each sub-region, obtain the first loss component, second loss component, and third loss component of the beam corresponding to shrinkage, creep, and prestressing tendon relaxation, respectively.

[0112] In this embodiment, three independent calculation modules are established in the calibrated finite element model: a concrete shrinkage calculation module, used to calculate concrete shrinkage strain according to the CEB-FIP model specification; a concrete creep calculation module, used to calculate concrete creep strain according to the CEB-FIP model specification; and a prestressed tendon relaxation calculation module, used to calculate stress relaxation loss based on the prestressed tendon material properties.

[0113] For each sub-region, three independent simulation calculations are performed: the first simulation activates only the concrete shrinkage calculation module, while keeping the creep and prestressed tendon relaxation calculation modules in an inactive state; the second simulation activates only the concrete creep calculation module, while keeping the shrinkage and prestressed tendon relaxation calculation modules in an inactive state; the third simulation activates only the prestressed tendon relaxation calculation module, while keeping the shrinkage and creep calculation modules in an inactive state.

[0114] Each independent simulation uses the same historical environmental parameters and historical load data as in step 3 as input, and calculates the loss components of each sub-region using the following formula:

[0115] For the first simulation:

[0116]

[0117] In the formula, Indicates the first Each sub-region at the current moment The loss component caused by concrete shrinkage;

[0118] For the second deduction:

[0119]

[0120] In the formula, Indicates the first Each sub-region at the current moment The loss component caused by creep;

[0121] For the third simulation:

[0122]

[0123] In the formula, Indicates the first Each sub-region at the current moment The loss component caused by prestressing tendon relaxation; The stress relaxation amount of the prestressing tendon is calculated based on the prestressing tendon material relaxation test data.

[0124] In the above independent deduction process Used to characterize the Each sub-region at the current moment The loss component caused by concrete shrinkage. The larger this value, the more severe the prestress loss caused by concrete shrinkage in this area, reflecting that the negative impact of concrete shrinkage deformation on the prestress retention capacity in this area is more significant. Used to characterize the Each sub-region at the current moment The increase in the value of the loss component caused by creep indicates that the creep deformation in this area is more severe under continuous load, resulting in more serious prestress loss, reflecting the cumulative effect of load history and time-varying material properties. Used to characterize the Each sub-region at the current moment The loss component caused by prestressing tendon relaxation means that the stress relaxation of the prestressing tendons under long-term stress is more obvious, which directly weakens the ability to maintain effective prestress. In summary, the magnitude of these three loss components directly reflects the contribution of their respective mechanisms to prestress loss. The larger the value, the higher the risk of loss in the area under the corresponding mechanism, providing a quantitative basis for identifying the dominant loss factors and formulating targeted maintenance measures.

[0125] The individual loss components of each sub-region are weighted and summed to obtain the first, second, and third loss components corresponding to the beam. The formulas used are as follows:

[0126]

[0127] In the formula, Indicates the current moment The first loss component;

[0128]

[0129] In the formula, Indicates the current moment The second loss component;

[0130]

[0131] In the formula, Indicates the current moment The third loss component.

[0132] By calculating the first, second, and third loss components, it is possible to achieve a refined tracing of the causes of prestress loss and a quantitative identification of the dominant factors. By decomposing the overall loss index into three independent loss components, it is possible not only to accurately reflect the contribution ratio of each time-varying mechanism to the total loss, but also to reveal the differences in response of different regions under shrinkage, creep, and tendon relaxation. This component-based quantification method allows the evaluation results to go beyond the macroscopic judgment of the overall loss value, and to deeply identify the key links that lead to prestress loss and their spatial distribution characteristics. This provides a targeted decision-making basis for subsequent structural maintenance, tension compensation, or reinforcement design, enabling accurate prediction and proactive control of the long-term performance of the structure.

[0133] Step 5: Based on the preset future environmental parameters and expected load, drive the calibrated model to extrapolate forward and predict the changing trends of the global comprehensive loss index, the first loss component, the second loss component and the third loss component at the future target time point.

[0134] In this embodiment, based on the calibrated finite element model that has completed historical extrapolation and loss component calculation, preset future environmental parameters and expected loads are used as input conditions; the future environmental parameters include the daily average temperature and daily average relative humidity of each day in the future target time period obtained based on meteorological forecast data; the expected loads include the maximum value of the equivalent uniformly distributed load of each day in the future target time period determined based on design usage requirements.

[0135] The calibrated model is then used for forward extrapolation calculations to predict the global comprehensive loss index at future target time points using the following formula:

[0136]

[0137] In the formula, For future target time points The predicted value of the overall loss index across the entire region; For the current moment The overall loss index of the entire domain; Indicates from the current moment To the future target time point The incremental loss within the period is calculated using the following formula:

[0138]

[0139] in, For concrete stress predicted based on expected load; The creep rate of concrete is predicted based on future environmental parameters; This refers to the concrete shrinkage rate predicted based on future environmental parameters.

[0140] In the formula for calculating the predicted value of the global comprehensive loss index, the dependent variable... Used to characterize the target time point in the future. The predicted value of the overall prestress loss index of the prestressed beam reflects the cumulative degree of overall prestress loss of the beam under the combined effect of time-varying effects such as concrete shrinkage, creep and prestressing tendon relaxation during the predicted time period. A higher value indicates a more severe loss of prestress in the beam structure in the future, a higher risk of overall structural performance degradation, and potential impact on its long-term load-bearing capacity and durability; conversely, a lower value indicates a lower prestress loss. The smaller the value, the better the prestress level of the beam can be maintained in the future, the overall performance of the structure is relatively stable, and the long-term safety status is relatively controllable.

[0141] Based on the loss component calculation method in step 4, the changing trends of the first, second, and third loss components at future target time points are predicted respectively. The specific logic is as follows:

[0142] First loss component prediction:

[0143]

[0144] In the formula, For future target time points The predicted value of the first loss component;

[0145] Second loss component prediction:

[0146]

[0147] In the formula, For future target time points The predicted value of the second loss component;

[0148] Third loss component prediction:

[0149]

[0150] In the formula, For future target time points The predicted value of the third loss component;

[0151] Setting up independent predictions for the first, second, and third loss components aims to further refine and causally-oriented predict future prestressed loss trends, building upon the existing regionalized historical loss decomposition. By decomposing the future prediction of the overall loss index into the superposition of three independent components—shrinkage, creep, and prestressing tendon relaxation—it not only continues the source-tracing analysis of loss mechanisms from step 4, but more importantly, it quantifies the contribution of each time-varying effect to loss growth and its evolution under future environmental and load conditions. This design allows the prediction results to clearly indicate which loss mechanism will dominate in the future, thus providing a more targeted basis for formulating structural maintenance strategies. For example, if the prediction shows a significant increase in creep loss, priority can be given to optimizing load management or strengthening creep monitoring; if shrinkage loss is dominant, the focus can be on controlling environmental humidity or implementing shrinkage compensation measures.

[0152] Based on the calculated future target time points The predicted value of the overall loss index First loss component prediction value Predicted value of the second loss component and the predicted value of the third loss component The changing trends of each parameter can be obtained in the following ways:

[0153] The current moment's overall loss index With future target time points The predicted value of the overall loss index By comparing and calculating their absolute changes, the growth rate of the overall loss index can be characterized.

[0154] The first, second, and third loss components at the current moment are compared with the future target time point. The predicted values ​​of the first, second, and third loss components are compared to calculate their absolute changes. By comparing the magnitude of the absolute changes of each loss component, the dominant factor contributing the most to the growth of total loss is identified. By setting multiple consecutive future target time points for repeated prediction, the overall loss index and the sequence data of each loss component changing over time are obtained. Based on this sequence data, the correspondence between time and loss value is established, and the trend function of each parameter is obtained by curve fitting, thereby quantifying its long-term development law.

[0155] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0156] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0157] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0158] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for evaluating the prestress loss of a prestressed beam based on digital deduction, characterized in that, The specific steps are as follows: Step 1: Construct a finite element model of the prestressed beam, and evenly divide the finite element model into multiple sub-regions along the length direction of the beam body. For each sub-region, based on its material parameters and historical environmental parameters, determine the regional sensitivity coefficient used to characterize the time-varying characteristics of the material in this sub-region; Step 2: Collect the initial strain and pressure data of each sub-region at the starting point of the time-step calculation, compare it with the initial simulation data of the model, and adjust the duct friction coefficient and anchor deformation value in the model through inverse analysis to complete model calibration; Step 3: Input the historical environmental parameters and historical load data of each sub-region from the starting point of the time-step calculation to the current moment into the calibrated model in a time series, drive the model to perform time-step calculations, simulate the structural responses of each sub-region under the corresponding historical environment and loads, and based on the structural responses, combined with the regional sensitivity coefficient, calculate the time-varying loss values of each sub-region, and summarize the time-varying loss values of all sub-regions to generate the global comprehensive loss index; Step 4: For each sub-region, perform independent deduction calculations considering only concrete shrinkage, only creep, and only relaxation of prestressed tendons respectively. By summarizing the independent deduction results of each sub-region, obtain the first loss component, the second loss component, and the third loss component of the beam body corresponding to shrinkage, creep, and relaxation of prestressed tendons respectively; Step 5: Based on the preset future environmental parameters and expected loads, drive the calibrated model to perform forward deduction, and predict the change trends of the global comprehensive loss index, the first loss component, the second loss component, and the third loss component at the future target time point.

2. The prestress loss assessment method of a prestressed beam based on digital deduction according to claim 1, characterized in that: The specific implementation process of Step 1 is as follows: According to the design and construction drawings of the prestressed beam, obtain the geometric dimensions of the beam body, the designed concrete strength grade, elastic modulus, the layout and material properties of prestressed steel tendons, and establish a parametric finite element model in finite element analysis software. Subsequently, evenly divide the finite element model into multiple sub-regions along the length direction of the beam body, and the division spacing is determined according to the length of the beam body; For each divided sub-region, collect its material parameters and historical environmental parameters. The material parameters include the actual compressive strength and volume reinforcement ratio of the concrete in this sub-region, where the actual compressive strength is obtained by the comprehensive method of ultrasonic rebound, and the volume reinforcement ratio is determined based on the design drawings and combined with the construction acceptance records; the historical environmental parameters refer to the daily average temperature and daily average relative humidity of each day in the recent 1 month in this sub-region, and are obtained through the historical monitoring records of the temperature and humidity sensors deployed at the corresponding positions of the beam body; For each sub-region, based on its material parameters and historical environmental parameters, determine the regional sensitivity coefficient used to characterize the time-varying characteristics of the material in this sub-region, which specifically includes: Obtain the actual compressive strength of the concrete in the sub-region and the reference value of the concrete compressive strength, and perform a weighted sum based on the ratio of the two, the volume reinforcement ratio, and the comprehensive index of temperature and humidity fluctuations. The comprehensive index of temperature and humidity fluctuations is the sum of the ratio of the standard deviation to the mean of the daily average temperature and the ratio of the standard deviation to the mean of the daily average relative humidity in the sub-region in the most recent month.

3. The prestress loss evaluation method of a prestressed beam based on digital deduction according to claim 2, wherein: Divide a time period that ends at the current moment and trace back one month in length from the current moment as the starting point of the time period, and mark the starting point of the time period as the starting point for time step calculation; at the starting point for time step calculation, for each sub-region, arrange pressure sensors at the anchorage ends of the prestressed tendons corresponding thereto to collect the pressure data under the anchorages, and at the same time arrange strain gauges at the mid-span section, quarter-point section and support section of each sub-region respectively to collect the concrete surface strain data; form the measured data set with the initial strain and pressure data of each sub-region at the starting point for time step calculation, and form the initial simulation data set with the corresponding data calculated by the finite element model under the same boundary conditions; Compare the measured data set with the initial simulation data set item by item, calculate the relative error of each data point, establish the objective function of inverse analysis with the minimization of the sum of the squares of the relative errors of all data points as the optimization goal, take the duct friction coefficient and anchor deformation value as the variables to be optimized, and use the gradient descent method for iterative optimization; In each iteration, update the finite element model parameters according to the current optimization variables, recalculate the initial simulation data set and evaluate the objective function value. Stop the optimization when the objective function value is less than the preset threshold. At this time, the obtained duct friction coefficient and the value of the anchorage deformation are the optimal solutions, and the calibration of the finite element model is completed.

4. A method for evaluating prestress loss of a prestressed beam based on digital deduction according to claim 1, characterized in that: The historical environmental parameters and historical load data collected from each sub-region from the starting point of time-step calculation to the current moment are formed into a complete time series in chronological order and input into the calibrated finite element model; the historical environmental parameters refer to the daily average temperature and daily average relative humidity of each sub-region from the starting point of time-step calculation to the current moment during this period; the historical load data is the maximum value of the daily equivalent uniform load acting on the beam body collected by the structural health monitoring system during the corresponding period. Drive the finite element model to perform time-step calculations according to the input time series, and simulate the structural responses of each sub-region at each time step. The structural responses include the stress time history and strain time history of the concrete. Based on the simulated structural responses and combined with the regional sensitivity coefficients calculated in Step 1, calculate the time-varying loss values of each sub-region, specifically including: Obtain the concrete stress of the sub-region at time according to the finite element simulation, and calculate the concrete creep rate and concrete shrinkage rate at the corresponding time according to the CEB-FIP model code. At the same time, combine the concrete elastic modulus obtained from the laboratory material test; based on the regional sensitivity coefficient, the product of the concrete stress and creep rate at time , and the product of the concrete elastic modulus and shrinkage rate, perform an integral operation within the time interval from the start point of the time step calculation to the current time to obtain the time-varying loss value of the sub-region at the current time ; where is the integral time variable; Weighted sum the time-varying loss values of each sub-region at the same moment \(t\) according to their respective volume ratios as weights to obtain the global comprehensive loss index at the current moment of the whole region.

5. The prestress loss assessment method of a prestressed beam based on digital deduction according to claim 4, wherein: Establish three independent calculation modules in the calibrated finite element model: a concrete shrinkage calculation module for calculating the concrete shrinkage strain according to the CEB-FIP model code; a concrete creep calculation module for calculating the concrete creep strain according to the CEB-FIP model code; a prestressed tendon relaxation calculation module for calculating the stress relaxation loss according to the material properties of the prestressed tendon. Perform three independent deduction calculations for each sub-region respectively: in the first deduction, only activate the concrete shrinkage calculation module and keep the creep and prestressed tendon relaxation calculation modules in the non-activated state; in the second deduction, only activate the concrete creep calculation module and keep the shrinkage and prestressed tendon relaxation calculation modules in the non-activated state; in the third deduction, only activate the prestressed tendon relaxation calculation module and keep the shrinkage and creep calculation modules in the non-activated state. Each independent deduction uses the same historical environmental parameters and historical load data as in Step 3 as inputs to calculate the loss components of each sub-region, specifically including: For the first deduction, calculate the loss component caused by concrete shrinkage: Based on the regional sensitivity coefficient of the sub-region, the elastic modulus of concrete, and the concrete shrinkage rate calculated according to the CEB-FIP model code, perform an integral operation within the time interval from the starting point of the time step calculation to the current moment to obtain the loss component caused by concrete shrinkage in this sub-region at the current moment ; For the second deduction, calculate the loss component caused by creep: Based on the regional sensitivity coefficient of the sub-region, the concrete stress of this region obtained through finite element simulation at time , and the concrete creep rate calculated according to the CEB-FIP model code, perform integral operation within the time interval from the starting point of the time step calculation to the current time to obtain the loss component caused by creep in this sub-region at the current time ; For the third deduction, calculate the loss component caused by the relaxation of the prestressing tendons: Based on the regional sensitivity coefficient of the sub-region and the amount of stress relaxation of the prestressing tendons calculated according to the relaxation test data of the prestressing tendon material, perform an integral operation within the time interval from the start point of the time step calculation to the current moment to obtain the loss component caused by the relaxation of the prestressing tendons in this sub-region at the current moment ; Perform weighted summation on the single loss components of each sub-region to obtain the corresponding first loss component, second loss component and third loss component of the beam body, specifically including: For each sub-region at the current moment The loss components caused by concrete shrinkage are weighted and summed according to their respective volume ratios as weights to obtain the first loss component of the beam body at the current moment ; At the current moment for each sub-region The loss components caused by creep are weighted and summed according to their respective volume ratios as weights to obtain the second loss component of the beam body at the current moment; At the current moment for each sub-region The loss components caused by the relaxation of the prestressing tendons are weighted and summed using their respective volume ratios as weights to obtain the third loss component of the beam body at the current moment 6. The method for evaluating the prestress loss of a prestressed beam based on digital deduction according to claim 5, wherein: Based on the calibrated finite element model that has completed historical deduction and loss component calculation, use the preset future environmental parameters and expected loads as input conditions. The future environmental parameters include the daily average temperature and daily average relative humidity of each day within the future target time period based on meteorological prediction data. The expected loads include the maximum value of the equivalent uniformly distributed load of each day within the future target time period determined based on the design use requirements. Drive the calibrated model to perform forward deduction calculations to predict the global comprehensive loss index at the future target time point, specifically including: Based on the overall comprehensive loss index at the current moment as the base value, and adding the increment of the loss index from the current moment to the future target time point, the predicted value of the overall comprehensive loss index at the future target time point is obtained; The calculation method of the loss index increment is: for each sub-region, take the product of its weight coefficient, the regional sensitivity coefficient of the sub-region, and an integral term as the contribution term of the sub-region. The integral term represents the result of integrating the product of the concrete stress predicted based on the expected load and the concrete creep rate predicted based on the future environmental parameters, plus the product of the concrete elastic modulus and the concrete shrinkage rate predicted based on the future environmental parameters, within the time interval from the current moment to the future target time point. Sum up the contribution terms of all sub-regions to obtain the loss index increment.

7. A method for evaluating prestress loss of prestressed beams based on digital deduction according to claim 6, characterized in that: Based on the loss component calculation method in Step 4, the changing trends of the first loss component, the second loss component, and the third loss component at a future target time point are predicted respectively, and the specific logic is as follows: The prediction method for the first loss component is: Taking the first loss component at the current moment as the base value, and adding the increment of the first loss component from the current moment to the future target time point, the predicted value of the first loss component at the future target time point is obtained; The calculation method for the increment of the first loss component is: for each sub-region, the product of its weight coefficient, the regional sensitivity coefficient of this sub-region, and an integral term is used as the contribution term of this sub-region, where the integral term represents the result of integrating the product of the concrete elastic modulus and the predicted concrete shrinkage rate based on future environmental parameters within the time interval from the current moment to the future target time point. Summing up the contribution terms of all sub-regions, the increment of the first loss component is obtained; The prediction method for the second loss component is: Taking the second loss component at the current moment as the base value, and adding the increment of the second loss component from the current moment to the future target time point, the predicted value of the second loss component at the future target time point is obtained; The calculation method for the increment of the second loss component is: for each sub-region, the product of its weight coefficient, the regional sensitivity coefficient of this sub-region, and an integral term is used as the contribution term of this sub-region, where the integral term represents the result of integrating the product of the predicted concrete stress based on the expected load and the predicted concrete creep rate based on future environmental parameters within the time interval from the current moment to the future target time point. Summing up the contribution terms of all sub-regions, the increment of the second loss component is obtained; The prediction method for the third loss component is: Taking the third loss component at the current moment as the base value, and adding the increment of the third loss component from the current moment to the future target time point, the predicted value of the third loss component at the future target time point is obtained; The calculation method for the increment of the third loss component is: for each sub-region, the product of its weight coefficient, the regional sensitivity coefficient of this sub-region, and an integral term is used as the contribution term of this sub-region, where the integral term represents the result of integrating the predicted prestress relaxation of the prestressing tendon based on the material properties of the prestressing tendon within the time interval from the current moment to the future target time point. Summing up the contribution terms of all sub-regions, the increment of the third loss component is obtained; Based on the predicted values of the global comprehensive loss index, the first loss component, the second loss component, and the third loss component at the future target time point obtained by calculation, the changing trends of each parameter are obtained through the following method: Compare the global comprehensive loss index at the current moment with the predicted value of the global comprehensive loss index at the future target time point, and calculate its absolute change amount to characterize the growth rate of the global comprehensive loss index; Compare the first loss component, the second loss component, and the third loss component at the current moment with the predicted values of the first loss component, the second loss component, and the third loss component at the future target time point respectively, and calculate their absolute change amounts; identify the dominant factor that contributes the most to the total loss growth by comparing the magnitudes of the absolute change amounts of each loss component; obtain the global comprehensive loss index and the sequence data of the change of each loss component over time by performing repeated predictions by setting multiple consecutive future target time points, establish the corresponding relationship between time and loss value based on this sequence data, use the curve fitting method to obtain the change trend function of each parameter, and further quantify its long-term development law.

Images