Intelligent road infrastructure health state prediction method

Abstract

The application discloses a smart road infrastructure health state prediction method, and relates to the field of intelligent transportation.The method comprises the following steps: in an edge computing node, a road strain distribution matrix is calculated based on a quantum entangled photon pair signal; a road digital twin is constructed by fusing road three-dimensional scanning information, road surface pressure distribution information and the road strain distribution matrix; a road structure health index is calculated according to the road digital twin; a predicted health decline rate is obtained by processing the road structure health index using a gated quantum recurrent neural network; and when the predicted health decline rate exceeds a preset rate threshold, a self-repairing microcapsule rupture instruction is triggered to release a bitumen regenerating agent to fill microcracks and perform automatic road repair. The method solves the technical problem of insufficient real-time performance and accuracy of existing smart road infrastructure health state prediction, and achieves the technical effect of improving the real-time performance and accuracy of road infrastructure health state prediction.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation, and in particular to a method for predicting the health status of smart road infrastructure. Background Technology

[0002] Predicting and managing the health status of intelligent road infrastructure is crucial for ensuring road safety, extending its lifespan, and improving traffic efficiency. Currently, the main approach to addressing this issue relies on traditional road inspection and maintenance methods, such as regular manual checks and sensor monitoring of road conditions. These methods combine data analysis to assess road health. However, current methods are limited by the real-time nature and accuracy of data processing, and lack sufficient sensitivity to minor damage and potential structural changes. This makes it difficult to detect and predict declining road health trends in a timely and effective manner, potentially leading to safety hazards and increased maintenance costs.

[0003] Currently, the prediction of the health status of smart road infrastructure suffers from technical problems such as insufficient real-time performance and accuracy. Summary of the Invention

[0004] This application provides a method for predicting the health status of intelligent road infrastructure. It employs techniques such as calculating the road strain matrix using quantum signals at the edge node, constructing a digital twin of the road by integrating 3D scanning and pressure distribution information, calculating a health index based on this, predicting the rate of health decline through a quantum neural network, and triggering a self-repair command to release regenerant to repair the road if the rate of decline exceeds the limit. These techniques achieve the technical effect of improving the real-time performance and accuracy of predicting the health status of road infrastructure.

[0005] This application provides a method for predicting the health status of smart road infrastructure, comprising: calculating a road strain distribution matrix based on quantum entangled photon pair signals at an edge computing node; constructing a road digital twin by fusing road 3D scanning information, road surface pressure distribution information, and the road strain distribution matrix; calculating a road structural health index based on the road digital twin; processing the road structural health index using a gated quantum recurrent neural network to obtain a predicted health degradation rate; and triggering a self-healing microcapsule rupture command when the predicted health degradation rate exceeds a preset rate threshold, releasing asphalt recycling agent to fill microcracks and automatically repairing the road.

[0006] In a possible implementation, at the edge computing node, based on quantum entangled photon pair signals, the road strain distribution matrix is ​​calculated, and the following processing is performed: an FBG array is deployed within the road asphalt layer, and a quantum entangled photon pair generator is deployed beside the road; the quantum entangled photon pair generator generates a first photon beam and a second photon beam, the first photon beam is injected into the FBG array to obtain a first photon signal, and the second photon beam is transmitted to the edge computing node to obtain a second photon signal; the road strain distribution matrix is ​​calculated based on the first photon signal and the second photon signal.

[0007] In a possible implementation, the road strain distribution matrix is ​​calculated based on the first photon signal and the second photon signal, and the following processing is performed: the coincidence count rate of the first photon signal and the second photon signal is measured, and the road strain distribution matrix is ​​calculated using the following formula: Where ∑ is the road strain distribution matrix, K is the material strain coefficient, ΔC is the coincidence count rate, ΔC0 is the reference coincidence count, α is the temperature decay factor, T is the asphalt layer temperature, and n is the empirical coefficient.

[0008] In a possible implementation, the construction of a road digital twin involves fusing road 3D scanning information, road surface pressure distribution information, and the road strain distribution matrix, and performing the following processes: representing the road 3D scanning information as a road 3D geometric model; representing the road surface pressure distribution information as a road pressure distribution map; representing the road strain distribution matrix as a road stress distribution map; fusing the road 3D geometric model, the road pressure distribution map, and the road stress distribution map to obtain a road comprehensive feature tensor; and constructing the road digital twin based on the road comprehensive feature tensor.

[0009] In a possible implementation, the process of calculating the road structure health index based on the road digital twin involves the following steps: collecting historical road health status data, including historical road comprehensive feature tensors and corresponding historical road structure health indices under known health states; supervising the training of a health index calculation model based on the historical road comprehensive feature tensors and the historical road structure health indices; extracting the current road comprehensive feature tensor from the road digital twin and inputting it into the health index calculation model to obtain the road structure health index.

[0010] In a possible implementation, the road structure health index is processed using a gated quantum recurrent neural network to obtain a predicted rate of health decline. The following processing is performed: collecting time-series data of historical road structure health index, including a sequence of historical road structure health index under known health conditions and a corresponding sequence of historical health decline rates; training the gated quantum recurrent neural network based on the historical road structure health index sequence and the historical health decline rate sequence; inputting a time-series input vector composed of the current road structure health index and the road structure health indices of the previous M time points into the trained gated quantum recurrent neural network, and calculating and outputting the predicted rate of health decline.

[0011] In a possible implementation, when the predicted rate of health decline exceeds a preset rate threshold, a self-healing microcapsule rupture command is triggered to release asphalt recycling agent to fill microcracks and perform automatic road repair. The following processing is also performed: establishing a meteorological-road health status correlation model based on the relationship between meteorological conditions and road health status; inputting real-time meteorological parameters into the meteorological-road health status correlation model to obtain the expected road health change coefficient; and dynamically adjusting the preset rate threshold based on the expected road health change coefficient.

[0012] In a possible implementation, the following processing is performed: the edge computing node includes a quantum entanglement strain calculation module, a road digital twin construction module, a health index calculation module, a health status prediction module, and a self-repair instruction triggering module.

[0013] The proposed method for predicting the health status of intelligent road infrastructure first calculates the road strain distribution matrix at an edge computing node based on quantum entangled photon pair signals. Then, by fusing road 3D scan information, road surface pressure distribution information, and the road strain distribution matrix, a digital twin of the road is constructed. Next, based on this digital twin, a road structural health index is calculated. This index is then processed using a gated quantum recurrent neural network to obtain the predicted health degradation rate. Finally, when the predicted health degradation rate exceeds a preset threshold, a self-healing microcapsule rupture command is triggered, releasing asphalt recycling agent to fill the microcracks and automatically repairing the road. This method achieves the technical effect of improving the real-time performance and accuracy of road infrastructure health status prediction. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the method according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0015] Figure 1 This is a flowchart illustrating the method for predicting the health status of smart road infrastructure provided in this application embodiment.

[0016] Figure 2 This is a schematic diagram illustrating the process of dynamically adjusting a preset rate threshold in the smart road infrastructure health status prediction method provided in this application embodiment. Detailed Implementation

[0017] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0020] This application provides a method for predicting the health status of smart road infrastructure, such as... Figure 1As shown, the method includes:

[0021] Step S100: At the edge computing node, calculate the road strain distribution matrix based on the quantum entangled photon pair signal.

[0022] Specifically, leveraging the properties of quantum entanglement, entangled photon pairs are generated through specialized quantum communication devices. One photon is sent to a road monitoring device, while the other remains at an edge computing node. When the photon on the road monitoring device is affected by road strain, its state changes, influencing the photon at the edge computing node through quantum entanglement. The edge computing node is a computing device deployed near the road to process data collected from the road monitoring device in real time. The road strain distribution matrix is ​​calculated using the output signal of the quantum communication device and a specific algorithm to determine the strain at various points on the road. For example, a road 100 meters long and 10 meters wide is divided into 1000 monitoring points. The strain data at each monitoring point is expressed in microstrain (με), ranging from 0 to 1000 με. The strain distribution matrix is ​​a 100x10 two-dimensional array, where each element represents the strain value at the corresponding monitoring point.

[0023] In one possible implementation, the calculation of the road strain distribution matrix based on quantum entangled photon pair signals at the edge computing node, step S100 further includes step S110: deploying an FBG array within the road asphalt layer and deploying a quantum entangled photon pair generator beside the road. Specifically, a fiber Bragg grating (FBG) is a device capable of reflecting light of a specific wavelength, the reflected wavelength of which is linearly related to the strain experienced by the FBG. Deploying an FBG array within the road asphalt layer involves embedding a series of FBGs at different locations along the road to monitor the strain at various points. These FBGs are connected by optical fibers to form a network capable of transmitting optical signals. Quantum entanglement is a phenomenon in quantum mechanics where two or more particles are correlated in such a way that the state of one particle can instantaneously affect the state of another particle, regardless of their distance. Deploying a quantum entangled photon pair generator beside the road generates a pair of entangled photons, namely a first photon beam and a second photon beam.

[0024] In step S120, the quantum entangled photon pair generator generates a first photon beam and a second photon beam. The first photon beam is injected into the FBG array to obtain a first photon signal, and the second photon beam is transmitted to the edge computing node to obtain a second photon signal. Specifically, the quantum entangled photon pair generator produces a pair of entangled photons, labeled as the first photon beam and the second photon beam, respectively. The first photon beam is injected into the FBG array deployed in the road asphalt layer through an optical fiber. Due to the strain sensitivity of the FBG, it changes the wavelength of the first photon beam according to the road strain, thereby generating a first photon signal containing road strain information. The second photon beam is directly transmitted to the edge computing node through another optical fiber as a reference signal.

[0025] Step S130: Based on the first photon signal and the second photon signal, a road strain distribution matrix is ​​calculated. Specifically, at the edge computing node, the quantum states of the first and second photon signals are analyzed using quantum measurement techniques (such as quantum state tomography) to extract road strain information. Using the extracted road strain information, combined with the layout of the FBG array, a road strain distribution matrix is ​​constructed. Each element of this matrix represents the strain value at the corresponding location on the road. This implementation utilizes the combination of FBG and quantum entanglement technology to achieve high-precision real-time monitoring of road strain, improving the accuracy of road health status prediction. The introduction of edge computing nodes makes data processing more efficient, enabling the calculation of the road strain distribution matrix in real time on-site, providing timely data support for subsequent prediction and repair.

[0026] In one possible implementation, the step S130, which calculates the road strain distribution matrix based on the first photon signal and the second photon signal, further includes step S131, which measures the coincidence count rate of the first photon signal and the second photon signal, and calculates the road strain distribution matrix using the following formula:

[0027]

[0028] Where ∑ is the road strain distribution matrix, K is the material strain coefficient, ΔC is the coincidence count rate, ΔC0 is the reference coincidence count, α is the temperature decay factor, T is the asphalt layer temperature, and n is the empirical coefficient.

[0029] Specifically, at the edge computing node, photon detectors are set up to receive a first photon signal (transmitted from the FBG array) and a second photon signal (transmitted directly from the quantum entangled photon pair generator). The photon detectors convert the photon signals into electrical signals and record the number of coincidence events using a counter. A coincidence event refers to an event where two detectors receive photons simultaneously or almost simultaneously within a certain time window. The coincidence count rate is obtained by repeatedly measuring and calculating the proportion of coincidence events to the total number of events. This ratio reflects the degree of quantum entanglement between the first and second photon signals, and also indirectly reflects road strain information. The road strain distribution matrix is ​​calculated using a given formula, where K is the material strain coefficient, a constant predetermined based on the characteristics of the road material. This coefficient is used to correlate the quantum measurement results with the actual road strain. The baseline coincidence count is the coincidence count rate measured without road strain (or with negligible strain), serving as a reference benchmark. The temperature decay factor is a function related to the asphalt layer temperature T, reflecting the influence of temperature on the quantum signal and road strain measurements. This function is obtained by fitting experimental data. T is the actual temperature of the asphalt layer, measured in real time by a temperature sensor. 'n' is an empirical coefficient used to adjust the influence of the temperature decay factor on the coincidence count rate; this coefficient is optimized through experimental data. This implementation utilizes the coincidence count rate of quantum entangled photon pairs to indirectly measure road strain, which can improve the accuracy and sensitivity of the measurement.

[0030] In one possible implementation, step S100 further includes step S140, wherein the edge computing node includes a quantum entangled strain calculation module, a road digital twin construction module, a health index calculation module, a health status prediction module, and a self-repair instruction triggering module.

[0031] Specifically, the quantum entangled strain calculation module is responsible for calculating the road strain distribution matrix. It receives the first and second photon signals transmitted from the quantum entangled photon pair generator and measures the coincidence count rate between them. The module integrates a photon detector, a counter, and a coincidence logic circuit. The photon detector converts the photon signals into electrical signals, the counter records the number of coincidence events, and the coincidence logic circuit determines whether the two detectors receive photons simultaneously based on a preset time window, thereby calculating the coincidence count rate.

[0032] The road digital twin construction module is responsible for fusing road 3D scan information, road surface pressure distribution information, and the road strain distribution matrix obtained by the quantum entanglement strain calculation module to construct a road digital twin. The module integrates a data fusion algorithm and 3D modeling software. The data fusion algorithm integrates data from different sources, while the 3D modeling software constructs a 3D model of the road, i.e., the digital twin, based on the fused data.

[0033] The health index calculation module is responsible for calculating the health index of road structures. This module integrates a health index calculation algorithm that comprehensively assesses the health status of the road based on information such as road geometry, material properties, and strain distribution from a digital twin, and provides a quantified health index.

[0034] The health status prediction module is responsible for predicting the rate of road health decline. Internally, it integrates a gated quantum recurrent neural network model and training algorithm. This neural network learns the relationship between historical health index data and the rate of health decline, thereby predicting the future rate of health decline. The module receives historical health index data as training samples and current health index data as prediction input.

[0035] The self-healing command trigger module is responsible for triggering a self-healing microcapsule rupture command when the predicted rate of health degradation exceeds a preset rate threshold, releasing asphalt recycling agent to fill microcracks. The module integrates comparison logic and command sending functions. The comparison logic compares the predicted rate of health degradation with the preset rate threshold; if the threshold is exceeded, the command sending function sends a rupture command to the self-healing system. This implementation, by dividing the edge computing node into multiple functional modules, allows for more efficient processing and analysis of road health status-related data. Each module is responsible for a specific task, improving both processing accuracy and speed while facilitating system maintenance and upgrades.

[0036] Step S200: By fusing road 3D scanning information, road surface pressure distribution information and the road strain distribution matrix, a road digital twin is constructed.

[0037] Specifically, the road is scanned using equipment such as a 3D laser scanner to acquire its 3D geometric information. Pressure sensors deployed on the road surface are used to obtain real-time pressure distribution data. The 3D scan information, pressure distribution information, and strain distribution matrix are then fused to form the foundational data for a digital twin of the road. Finally, visualization or virtual reality technologies are used to construct a virtual environment for the digital twin.

[0038] In one possible implementation, by fusing road 3D scan information, road surface pressure distribution information, and the road strain distribution matrix to construct a road digital twin, step S200 further includes step S210, representing the road 3D scan information as a road 3D geometric model. Specifically, a high-precision 3D scanner is used to scan the road to obtain point cloud data of the road surface. Then, 3D modeling software (such as AutoCAD, Blender, etc.) is used to process the point cloud data to construct a 3D geometric model of the road, which accurately reflects the shape, size, and geometric features of the road.

[0039] Step S220: The road surface pressure distribution information is represented as a road pressure distribution map. Specifically, an array of pressure sensors is deployed on the road surface to monitor the pressure distribution in real time. The pressure data collected by the sensors is converted into a two-dimensional image, namely the road pressure distribution map, which visually displays the pressure magnitude at different locations on the road surface.

[0040] Step S230: The road strain distribution matrix is ​​represented as a road stress distribution map. Specifically, the road strain distribution matrix obtained by calculating the signal of quantum entangled photon pairs is converted into a road stress distribution map through color mapping or contour maps, etc. This image shows the strain or stress state at different locations inside the road.

[0041] Step S240: The road 3D geometric model, the road pressure distribution map, and the road stress distribution map are fused to obtain a comprehensive road feature tensor. Specifically, tensor fusion technology is used to fuse the road 3D geometric model, the road pressure distribution map, and the road stress distribution map. First, these three data sources are converted into tensors of the same size and resolution. Then, using methods such as tensor decomposition, tensor multiplication, or tensor concatenation, they are fused into a comprehensive feature tensor, which contains comprehensive information such as the road's 3D geometric features, surface pressure distribution, and internal stress state.

[0042] Step S250: Construct a digital twin of the road based on the road comprehensive feature tensor. Specifically, using 3D modeling software and simulation technology, a digital twin is constructed using the road comprehensive feature tensor. This digital twin is a virtual road model containing information such as all geometric features, surface pressure distribution, and internal stress state of the road. By updating the road comprehensive feature tensor in real time, the consistency between the digital twin and the real road condition is maintained. The digital twin allows for real-time monitoring and prediction of the road's health status. This implementation method improves the accuracy and reliability of road health status monitoring by integrating information from multiple data sources.

[0043] Step S300: Calculate the road structure health index based on the road digital twin.

[0044] Specifically, algorithms or software modules are used to process and analyze data from the road digital twin. Based on the analysis results, a road structural health index is calculated. The road structural health index is a comprehensive indicator reflecting the health status of the road structure, calculated based on geometric, mechanical, and damage information from the road digital twin.

[0045] In one possible implementation, the step of calculating the road structural health index based on the road digital twin further includes step S310, which involves collecting historical road health status data, including historical road comprehensive feature tensors and corresponding historical road structural health indices under known health conditions. Specifically, a large amount of historical road health status data is collected, including road comprehensive feature tensors and corresponding historical road structural health indices under known health conditions. Historical data can originate from long-term records of road monitoring systems. The collected data needs to be organized into a structured format, where the road comprehensive feature tensor contains the road's three-dimensional geometric information, pressure distribution information, and stress distribution information. This information is extracted and integrated from the raw data using the same methods as in steps S210-S250. The historical road structural health index is a quantitative assessment of road health status, labeled by road maintenance experts based on the actual condition of the road. For example, there is a dataset containing 1000 historical records, each containing a road comprehensive feature tensor and a corresponding road structural health index. These feature tensors are four-dimensional arrays, where three dimensions represent the road's length, width, and various physical quantities (such as geometric height, pressure, and strain), respectively. The fourth dimension stores the measurements of these physical quantities at different times or locations. The road structural health index is a floating-point number between 0 and 1, representing the road's health condition.

[0046] Step S320: Supervised training of a health index calculation model is performed based on the historical road comprehensive feature tensor and the historical road structural health index. Specifically, a health index calculation model is trained using collected historical data. The goal of this model is to predict the road structural health index based on the input road comprehensive feature tensor. Deep learning models, such as convolutional neural networks (CNNs) or recurrent neural networks (RNNs), can be selected to handle this spatially and temporally dependent data. The road comprehensive feature tensor from the historical dataset is used as input, and the historical road structural health index is used as output to train the model. During training, the model parameters are adjusted by minimizing the error between the predicted and actual values. The model's performance is evaluated using methods such as cross-validation, and the model architecture or parameters are adjusted as needed to optimize the prediction results.

[0047] Step S330: Extract the current comprehensive road feature tensor from the road digital twin and input it into the health index calculation model to obtain the road structure health index. Specifically, through a specific algorithm or interface, the current comprehensive road feature tensor is extracted from the real-time constructed road digital twin and input into the pre-trained health index calculation model. The model predicts and outputs the current road structure health index based on the learned knowledge. This implementation improves the efficiency and accuracy of road structure health index calculation by collecting historical data, supervising the training of the health index calculation model, and extracting features from the road digital twin for prediction.

[0048] Step S400: The road structure health index is processed using a gated quantum recurrent neural network to obtain the predicted rate of health decline.

[0049] Specifically, a gated quantum recurrent neural network (GRNNN) is a deep learning model that combines the characteristics of quantum computing and recurrent neural networks to process time-series data and predict future trends. A GRNNN model is constructed using a quantum machine learning framework or deep learning software. The road structure health index is used as input data to train the model to predict the future rate of health decline.

[0050] In one possible implementation, the process of using a gated quantum recurrent neural network to process the road structure health index to obtain the predicted rate of health decline further includes step S410, which involves collecting time-series data of historical road structure health indices, including a sequence of historical road structure health indices under known health conditions and the corresponding historical rate of health decline sequences. Specifically, a large amount of time-series data of historical road structure health indices is collected through a road monitoring system. This data includes a sequence of historical road structure health indices under known health conditions and the corresponding historical rate of health decline sequences. The historical road structure health index sequence is a time series, where each time point corresponds to a road structure health index. The historical rate of health decline sequence is calculated based on these health indices and represents the rate of change of road health status over time.

[0051] Step S420: Train a gated quantum recurrent neural network (GRNN) based on the historical road structure health index sequence and the historical health decline rate sequence. Specifically, a GRNN is trained using collected historical data. The goal of this network is to predict future health decline rates based on the input road structure health index time series. The GRNN combines the advantages of quantum computing and recurrent neural networks, exhibiting higher efficiency and accuracy when processing time series data. The specific architecture of the model includes an input layer, a quantum layer, a recurrent layer, and an output layer. The GRNN is trained using the historical road structure health index sequence as input and the historical health decline rate sequence as output. During training, the model parameters are optimized using the quantum gradient descent method. A cross-validation strategy is used to avoid overfitting, and the model's hyperparameters are adjusted based on the performance on the validation set.

[0052] Step S430 involves combining the current road structure health index and the road structure health indices of the previous M time points into a time series input vector, which is then input into a trained gated quantum recurrent neural network to calculate and output the predicted health decline rate. Specifically, based on the current time point, the current road structure health index is obtained from the road monitoring system and combined with the health indices of the previous M time points to form an input vector. This vector is then input into the gated quantum recurrent neural network for prediction. The model predicts and outputs the future health decline rate based on the learned knowledge. This implementation improves the accuracy and efficiency of prediction by collecting historical data, training the gated quantum recurrent neural network model, and inputting current data in real time for prediction.

[0053] Step S500: When the predicted rate of health decline exceeds a preset rate threshold, a self-healing microcapsule rupture command is triggered, releasing asphalt recycling agent to fill microcracks and perform automatic road repair.

[0054] Specifically, the system monitors and predicts the rate of road health degradation and compares it with a preset rate threshold (a preset threshold used to determine whether the rate of road health degradation has reached a level requiring automatic repair). When the predicted rate of health degradation exceeds the preset rate threshold, a trigger command is sent to the control system of the self-healing microcapsule via wireless communication technology, triggering its rupture and releasing asphalt rejuvenator. The self-healing microcapsule is a microcapsule containing asphalt rejuvenator embedded in the road material. When microcracks appear in the road, the microcapsule ruptures and releases asphalt rejuvenator to fill the cracks. This application's embodiment employs techniques such as calculating the road strain matrix using quantum signals at edge nodes, fusing 3D scanning and pressure distribution information to construct a digital twin of the road, calculating a health index based on this, predicting the rate of health degradation through a quantum neural network, and triggering a self-healing command to release rejuvenator to repair the road when the rate exceeds the threshold. These techniques achieve the technical effect of improving the real-time performance and accuracy of predicting the health status of road infrastructure.

[0055] like Figure 2 As shown, in one possible implementation, when the predicted rate of health decline exceeds a preset rate threshold, a self-healing microcapsule rupture command is triggered, releasing asphalt recycling agent to fill microcracks for automatic road repair. Step S500 further includes step S510, establishing a meteorological-road health status correlation model based on the relationship between meteorological conditions and road health status. Specifically, a large amount of historical data is collected, including road health monitoring data (such as crack width, depth, distribution, pavement smoothness, material strength, etc.) and meteorological data from the same period (such as temperature, humidity, rainfall, wind speed, etc.). The collected data is preprocessed to extract key features, such as the daily average, maximum, and minimum values ​​of meteorological parameters, as well as relevant indicators of road health status. A machine learning algorithm (such as random forest) is used to train the model, with the input being the feature-engineered meteorological data and road health status data, and the output being the trend or predicted value of road health status. The model performance is evaluated using methods such as cross-validation, and the model parameters are adjusted and optimized based on the evaluation results.

[0056] Step S520: Input real-time meteorological parameters into the meteorological-road health status association model to obtain the expected change coefficient of road health. Specifically, real-time meteorological parameters, including temperature, humidity, and rainfall, are obtained through meteorological monitoring stations or meteorological APIs. These real-time meteorological parameters are then input into the trained meteorological-road health status association model to obtain the expected change coefficient of road health. The expected change coefficient of road health represents the expected degree or rate of change in road health status under specific meteorological conditions.

[0057] Step S530: Dynamically adjust the preset rate threshold based on the expected road health change coefficient. Specifically, an adjustment coefficient is calculated based on the expected road health change coefficient to adjust the preset rate threshold. For example, if the road health condition is expected to deteriorate significantly (e.g., crack propagation accelerates under high temperature and dry conditions), the adjustment coefficient is increased to lower the preset rate threshold, making it easier to trigger automatic repair. The preset rate threshold is multiplied by the adjustment coefficient to obtain a new threshold. The new threshold is compared with the predicted rate of health decline. If the predicted rate of health decline exceeds the new threshold, a self-healing microcapsule rupture command is triggered. This implementation, by considering the impact of meteorological conditions on road health condition, can more accurately predict the trend of road health condition changes, thereby improving the triggering accuracy of automatic repair.

[0058] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for predicting the health status of intelligent road infrastructure, characterized in that, include: At the edge computing node, the road strain distribution matrix is ​​calculated based on the quantum entangled photon pair signal; A digital twin of the road is constructed by integrating road 3D scanning information, road surface pressure distribution information, and the road strain distribution matrix. Calculate the road structure health index based on the road digital twin; The road structure health index is processed using a gated quantum recurrent neural network to obtain the predicted rate of health decline; When the predicted rate of health decline exceeds a preset rate threshold, a self-healing microcapsule rupture command is triggered, releasing asphalt recycling agent to fill the microcracks and automatically repairing the road. The calculation of the road strain distribution matrix at the edge computing node, based on quantum entangled photon pair signals, includes: Deploy FBG arrays within the asphalt layer of the road and deploy quantum entangled photon pair generators beside the road; The quantum entangled photon pair generator generates a first photon beam and a second photon beam. The first photon beam is injected into the FBG array to obtain a first photon signal, and the second photon beam is transmitted to the edge computing node to obtain a second photon signal. The road strain distribution matrix is ​​calculated based on the first photon signal and the second photon signal; The step of calculating the road strain distribution matrix based on the first photon signal and the second photon signal includes: The coincidence count rate between the first photon signal and the second photon signal is measured, and the road strain distribution matrix is ​​calculated using the following formula: ∑=K ； Where ∑ is the road strain distribution matrix, and K is the material strain coefficient. It is the coincidence count rate, It is a baseline conformity count. is the temperature decay factor, where T is the temperature of the asphalt layer and n is an empirical coefficient; The process of constructing a digital twin of the road by fusing road 3D scanning information, road surface pressure distribution information, and the road strain distribution matrix includes: Represent the 3D scanning information of the road as a 3D geometric model of the road; The road surface pressure distribution information is represented as a road pressure distribution map; The road strain distribution matrix is ​​represented as a road stress distribution map; The road comprehensive feature tensor is obtained by integrating the road three-dimensional geometric model, the road pressure distribution map, and the road stress distribution map; A digital twin of the road is constructed based on the comprehensive feature tensor of the road.

2. The method for predicting the health status of intelligent road infrastructure as described in claim 1, characterized in that, The calculation of the road structure health index based on the road digital twin includes: Collect historical road health status data, including the comprehensive feature tensor of historical roads under known health status and the corresponding historical road structural health index; Based on the historical road comprehensive feature tensor and the historical road structure health index, a health index calculation model is trained under supervision. The road structure health index is obtained by extracting the current comprehensive feature tensor of the road from the road digital twin and inputting it into the health index calculation model.

3. The method for predicting the health status of intelligent road infrastructure as described in claim 1, characterized in that, The process of using a gated quantum recurrent neural network to process the road structure health index and obtain the predicted rate of health decline includes: Collect time series data of historical road structure health index, including historical road structure health index sequences under known health conditions and corresponding historical health decline rate sequences; A gated quantum recurrent neural network is trained based on the historical road structure health index sequence and the historical health decline rate sequence. The current road structure health index and the road structure health indices of the previous M time points are combined to form a time series input vector, which is then input into a trained gated quantum recurrent neural network to calculate and output the predicted health decline rate.

4. The method for predicting the health status of intelligent road infrastructure as described in claim 1, characterized in that, The step of triggering a self-healing microcapsule rupture command when the predicted rate of health decline exceeds a preset rate threshold, releasing asphalt recycling agent to fill microcracks and perform automatic road repair, further includes: Based on the correlation between meteorological conditions and road health status, a meteorological-road health status correlation model is established; Input real-time meteorological parameters into the meteorological-road health status association model to obtain the expected change coefficient of road health; The preset rate threshold is dynamically adjusted based on the expected change coefficient of road health.

5. The method for predicting the health status of intelligent road infrastructure as described in claim 1, characterized in that, The edge computing node includes a quantum entanglement strain calculation module, a road digital twin construction module, a health index calculation module, a health status prediction module, and a self-repair command triggering module.

Images