Intelligent sensing method and system for road network maintenance

Abstract

The invention provides an intelligent sensing method and system for road network maintenance, and relates to the field of intelligent operation and maintenance and road management, and the method comprises the steps: obtaining historical weather data and road flow data of a target road in a preset historical time zone through the monitoring of the Internet of Things; obtaining pavement features and service life of the target road; based on machine learning, performing operation and maintenance period correction analysis according to the historical weather data, the road flow data, the pavement features and the service life, and obtaining an optimized operation and maintenance period; and executing road operation and maintenance in the future time zone according to the optimized operation and maintenance period. The objective of the invention is to solve the technical problems that the conventional road maintenance is generally based on a fixed time interval, the operation and maintenance period cannot be dynamically adjusted, the resources are wasted, the maintenance is not timely and the like, and the operation and maintenance period is corrected and optimized by combining historical weather data, road flow data, pavement features and the service life and utilizing a machine learning model. And a more accurate and personalized operation and maintenance plan can be realized.

Description

Technical Field

[0001] The present invention relates to the fields of intelligent operation and maintenance and road management, and particularly to an intelligent perception method and system for road network maintenance. Background Art

[0002] Traditional road operation and maintenance management mostly relies on pre-determined cycles, usually arranging maintenance work annually or seasonally. This method ignores the actual road usage conditions and environmental changes, such as factors like weather, traffic flow, and road wear degree. This means that the actual damage processes of different roads vary greatly, and a unified maintenance cycle may lead to premature or delayed maintenance of some roads, increasing unnecessary maintenance costs and risks. For example, on urban roads with high traffic flow, frequent heavy pressure and environmental factors may cause the road to be damaged prematurely, and when performing pre-determined periodic maintenance on it, the best repair time may have been missed, affecting the normal use of the road and increasing the repair cost. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent perception method and system for road network maintenance to solve the technical problems that traditional road maintenance is usually based on fixed time intervals, unable to dynamically adjust the operation and maintenance cycle, resulting in resource waste, untimely maintenance, etc., including:

[0004] In a first aspect, the present invention provides an intelligent perception method for road network maintenance, including: obtaining historical weather data and road traffic flow data of a target road in a preset historical time zone through Internet of Things monitoring; obtaining the road surface characteristics and service life of the target road; based on machine learning, performing operation and maintenance cycle correction analysis according to the historical weather data, road traffic flow data, road surface characteristics, and service life to obtain an optimized operation and maintenance cycle; and performing road operation and maintenance inspection in a future time zone according to the optimized operation and maintenance cycle.

[0005] Preferably, the intelligent perception method for road network maintenance further includes: obtaining historical weather data of a target road in a preset historical time zone through Internet of Things monitoring, where the historical weather data includes average temperature, maximum temperature difference, and rainfall; obtaining road traffic flow data of a target road in a preset historical time zone through Internet of Things monitoring, where the road traffic flow data includes daily average vehicle flow and daily average large vehicle flow of the road.

[0006] Preferably, the intelligent perception method for road network maintenance further includes: obtaining the road surface characteristics of the target road, where the road surface characteristics include road surface type, road surface material, and road surface damage ratio; obtaining the service durations of multiple sections of the target road and calculating the mean value of the multiple service durations to obtain the service life.

[0007] Preferably, the intelligent perception method for road network maintenance further includes: constructing a cycle correction analysis model based on machine learning, and collecting sample data to train the cycle correction analysis model until convergence; inputting the historical weather data, road traffic data, road surface characteristics, and service life into the cycle correction analysis model for analysis, and outputting a cycle correction coefficient; performing operation and maintenance cycle correction analysis according to the cycle correction coefficient to obtain an optimized operation and maintenance cycle.

[0008] Preferably, the intelligent perception method for road network maintenance further includes: constructing a cycle correction analysis model based on a BP neural network, wherein the cycle correction analysis model includes Q correction analysis plugins; collecting a sample weather data set, a sample road traffic set, a sample road surface characteristic set, and a sample service life set, and statistically calculating the mean value of the historical cycle correction coefficients under different sample weather data, sample road traffic, sample road surface characteristics, and sample service life, denoted as the sample correction coefficient, to obtain a sample correction coefficient set; using the sample weather data set, the sample road traffic set, the sample road surface characteristic set, the sample service life set, and the sample correction coefficient set as training data to perform supervised training on the Q correction analysis plugins until convergence, to obtain the cycle correction analysis model.

[0009] Preferably, the intelligent perception method for road network maintenance further includes: using the sample weather data set, the sample road traffic set, the sample road surface characteristic set, the sample service life set, and the sample correction coefficient set as training data, and equally dividing the training data into Q parts to obtain Q training sets; using the Q training sets to perform supervised training on the Q correction analysis plugins until convergence, to obtain Q converged correction analysis plugins; integrating and constructing the cycle correction analysis model according to the Q converged correction analysis plugins, wherein the output of the cycle correction analysis model is the mean value of the outputs of the Q converged correction analysis plugins.

[0010] Preferably, the intelligent perception method for road network maintenance further includes: obtaining a predetermined operation and maintenance cycle; multiplying the cycle correction coefficient by the predetermined operation and maintenance cycle, and taking the product of the two as the optimized operation and maintenance cycle.

[0011] In a second aspect, the present invention further provides an intelligent perception system for road network maintenance, which is used to execute an intelligent perception method for road network maintenance as described in the first aspect, including: an associated information acquisition module, which is used to obtain historical weather data and road traffic data of a target road in a preset historical time period through Internet of Things monitoring; a road surface information acquisition module, which is used to obtain the road surface characteristics and service life of the target road; an operation and maintenance cycle correction analysis module, which is used to perform operation and maintenance cycle correction analysis based on machine learning according to the historical weather data, road traffic data, road surface characteristics and service life to obtain an optimized operation and maintenance cycle; and a road operation and maintenance inspection module, which is used to execute road operation and maintenance inspection in a future time period according to the optimized operation and maintenance cycle.

[0012] The embodiments of the present invention have the following advantages:

[0013] Obtain historical weather data and road traffic data of a target road in a preset historical time period through Internet of Things monitoring; obtain the road surface characteristics and service life of the target road; perform operation and maintenance cycle correction analysis based on machine learning according to the historical weather data, road traffic data, road surface characteristics and service life to obtain an optimized operation and maintenance cycle; execute road operation and maintenance inspection in a future time period according to the optimized operation and maintenance cycle; that is to say, by combining historical weather data, road traffic data, road surface characteristics and service life, and using a machine learning model to correct and optimize the operation and maintenance cycle, a more accurate and personalized operation and maintenance plan can be realized. Description of the Drawings

[0014] Figure 1 It is a flowchart of the steps of an intelligent perception method for road network maintenance according to the present invention;

[0015] Figure 2 It is a schematic structural diagram of an intelligent perception system for road network maintenance according to the present invention.

[0016] Description of the Reference Numerals:

[0017] The associated information acquisition module 10, the road surface information acquisition module 20, the operation and maintenance cycle correction analysis module 30, and the road operation and maintenance inspection module 40. Detailed Embodiments

[0018] By providing an intelligent perception method and system for road network maintenance, the present invention solves the technical problems that traditional road maintenance is usually based on fixed time intervals, cannot dynamically adjust the operation and maintenance cycle, and there are problems such as resource waste and untimely maintenance. By combining historical weather data, road traffic data, road surface characteristics and service life, and using a machine learning model to correct and optimize the operation and maintenance cycle, a more accurate and personalized operation and maintenance plan can be realized.

[0019] Next, the technical solutions in the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments of the present invention. It should be understood that the present invention is not limited by the exemplary embodiments described herein. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention. Additionally, it should be noted that for the sake of description, only parts related to the present invention are shown in the drawings rather than all of them.

[0020] Embodiment 1. Please refer to the attached Figure 1 drawings. The present invention provides an intelligent perception method for road network maintenance, which is applied to an intelligent perception system for road network maintenance, and specifically includes the following steps:

[0021] Step 101: Obtain historical weather data and road traffic flow data of the target road in a preset historical time period through Internet of Things monitoring.

[0022] Furthermore, the present invention further includes the following steps:

[0023] Obtain historical weather data of the target road in a preset historical time period through Internet of Things monitoring, where the historical weather data includes average temperature, maximum temperature difference, and rainfall; obtain road traffic flow data of the target road in a preset historical time period through Internet of Things monitoring, where the road traffic flow data includes average daily traffic volume and average daily large vehicle traffic volume.

[0024] Specifically, historical weather data of the area where the target road is located is monitored through Internet of Things technology. These data are used to reflect the climate conditions of the road environment. The historical weather data includes average temperature, maximum temperature difference, and rainfall. The average temperature refers to the average temperature in a preset historical time period (such as the past week, month, or quarter). Temperature has an important impact on the expansion and contraction of road surface materials. Especially under extreme climate conditions (such as high temperature or low temperature), it may cause road surface cracks or deformation. The maximum temperature difference refers to the maximum value of the temperature difference between day and night within this time period. In areas with large temperature differences (such as areas with large day-night temperature differences), it will accelerate the aging of road surface materials and the expansion of cracks. Therefore, understanding the temperature difference helps to evaluate the potential risk of road damage. Rainfall refers to the total precipitation or average daily precipitation in the historical time period. Rainfall affects the slipperiness of the road surface. Excessive precipitation may cause waterlogging, potholes, road surface cracks, etc. A large amount of precipitation or a continuous wet environment will accelerate the damage of asphalt or concrete road surfaces.

[0025] Monitor the traffic conditions on the road through Internet of Things technology and record the road traffic flow data. These data help analyze the usage and damage degree of the road. The road traffic flow data includes the average daily traffic volume and the average daily large vehicle traffic volume of the road. The average daily traffic volume of the road refers to the average value of the daily vehicle passing volume of the target road within a preset historical time period. The greater the traffic volume, the greater the force on the road surface and the more serious the wear degree. Especially during the traffic peak period, it may exacerbate the damage such as compression and cracks on the road surface. The average daily large vehicle traffic volume refers to the number of large vehicles (such as trucks, buses, etc.) passing through the road every day. The weight of large vehicles is much greater than that of ordinary small vehicles, and the oppression and wear degree on the road surface are also greater. The level of the average daily large vehicle traffic volume can help evaluate the bearing pressure and potential damage risk of the road.

[0026] The combination of these historical weather data and road traffic flow data helps to intelligently analyze and predict the health status of the road. Factors such as temperature, precipitation, and traffic volume directly affect the service life and maintenance requirements of the road. Especially under the conditions of high traffic volume and extreme weather, more cracks, potholes, and aging problems may occur on the road surface. Therefore, by obtaining these data and conducting dynamic analysis, the road maintenance cycle can be optimized more precisely, ensuring the timely repair of the road, reducing maintenance costs, and improving the usage efficiency of the road network.

[0027] Step 102: Obtain the road surface characteristics and service life of the target road.

[0028] Furthermore, the present invention further includes the following steps:

[0029] Obtain the road surface characteristics of the target road, wherein the road surface characteristics include road surface type, road surface material, and road surface damage ratio; obtain the multiple usage durations of multiple sections of the target road, and calculate the average value of the multiple usage durations to obtain the service life.

[0030] Specifically, obtain the road surface characteristics of the target road, wherein the road surface characteristics include road surface type, road surface material, and road surface damage ratio. The road surface type refers to different material categories used on the road surface. Common road surface types include asphalt roads, concrete roads, etc.; the road surface material refers to the specific material type that constitutes the road surface. Different road surface materials have different durability, bearing capacity, and maintenance requirements. For example, materials such as asphalt mixtures, fine-grained concrete, and modified asphalt will affect the overall quality and maintenance cycle of the road; the road surface damage ratio refers to the degree of damage to the road surface, usually expressed by the ratio of the damaged area to the total road surface area.

[0031] The target road usually consists of multiple sections, and the service life of each section may be different. By recording the usage duration of each section, the actual service life of each section from the start of construction to the present can be obtained. To comprehensively evaluate the overall service life of the target road, the service lives of each section can be averaged. This average represents the average service life of the target road and provides a basis for subsequent maintenance decisions. For example, if the service life of some sections is long while that of other sections is short, the average can effectively reflect the service life of the entire road. Obtaining the road surface characteristics and service life provides more scientific data support for road maintenance, which can significantly improve the accuracy and intelligence level of road management.

[0032] Step 103: Based on machine learning, perform operation and maintenance cycle correction analysis according to the historical weather data, road traffic data, road surface characteristics, and service life to obtain an optimized operation and maintenance cycle.

[0033] Furthermore, the present invention further includes the following steps:

[0034] Construct a cycle correction analysis model based on machine learning, and collect sample data to train the cycle correction analysis model until it converges; input the historical weather data, road traffic data, road surface characteristics, and service life into the cycle correction analysis model for analysis, and output a cycle correction coefficient; perform operation and maintenance cycle correction analysis according to the cycle correction coefficient to obtain an optimized operation and maintenance cycle.

[0035] Specifically, design a machine learning model based on the multi-dimensional data of the target road (such as historical weather data, road traffic data, road surface characteristics, service life, etc.). The goal of this model is to identify the impact of different factors on road damage and maintenance cycles through learning historical data and predict the future maintenance needs of the road. By collecting sample data (including historical weather data, road traffic, road surface characteristics, and service life, etc.), and using it as a training set to input into the machine learning model for training. The training process is to repeatedly optimize the model parameters until the model obtains a high prediction accuracy on the test set, that is, convergence. Model convergence means that the model has been able to capture the complex relationships between data more accurately and make effective predictions.

[0036] Next, input the historical weather data, road traffic flow data, road surface characteristics, and service life into the periodic correction analysis model for analysis, and output the periodic correction coefficient. This coefficient represents the maintenance period that should be adjusted according to historical data and the current road conditions. Through the periodic correction coefficient, the operation and maintenance period of the traditional fixed time interval can be dynamically adjusted. This corrected period takes into account the impacts of factors such as weather changes, traffic flow, and road surface characteristics, making the operation and maintenance period more in line with the actual road usage conditions and damage trends. Finally, perform operation and maintenance period correction analysis based on the periodic correction coefficient to obtain the optimized operation and maintenance period. Among them, the corrected operation and maintenance period will be more adaptable. By comparing with the traditional regular maintenance period, the optimized period can reduce unnecessary early or delayed repairs on the premise of ensuring road safety, thereby reducing resource waste and improving maintenance efficiency.

[0037] This process applies a machine learning model to the periodic correction of road operation and maintenance. By collecting multi-dimensional data for model training, and then optimizing the operation and maintenance period, this not only improves the accuracy of road maintenance, but also enhances the intelligence and automation level of road network management, thereby effectively reducing costs and improving the usage safety of roads.

[0038] Furthermore, the present invention further includes the following steps:

[0039] Construct a periodic correction analysis model based on a BP neural network. Among them, the periodic correction analysis model includes Q correction analysis plugins; collect a sample weather data set, a sample road traffic flow set, a sample road surface characteristics set, and a sample service life set, and statistically calculate the mean of the historical periodic correction coefficients under different sample weather data, sample road traffic flow, sample road surface characteristics, and sample service life, denoted as the sample correction coefficient, to obtain a sample correction coefficient set; use the sample weather data set, sample road traffic flow set, sample road surface characteristics set, sample service life set, and sample correction coefficient set as training data to perform supervised training on the Q correction analysis plugins until convergence to obtain the periodic correction analysis model.

[0040] Specifically, the BP neural network is a commonly used supervised learning algorithm, especially suitable for scenarios that need to be trained through the mapping relationship between input and output data. In this model, the task of the BP neural network is to predict the maintenance period correction coefficient of the road based on the input multi-dimensional data (such as weather, traffic flow, road surface characteristics, etc.). Then construct a periodic correction analysis model based on the BP neural network. Among them, the periodic correction analysis model includes Q correction analysis plugins.

[0041] Then, collect the sample weather data set, sample road traffic volume set, sample road surface feature set, and sample service life set, and calculate the mean historical cycle correction coefficient under different sample weather data, sample road traffic volume, sample road surface features, and sample service life, which is set as the sample correction coefficient. That is, for each set of sample weather data, sample road traffic volume data, sample road surface feature data, and sample service life data, first analyze and calculate the historical cycle correction coefficient. This coefficient reflects whether the historical maintenance cycle needs to be adjusted under the given conditions. Then, calculate the mean historical cycle correction coefficient under different sample data, which is set as the sample correction coefficient. These coefficients provide a target output value for the model, that is, when these data are input, the model should predict the adjustment amount of the operation and maintenance cycle, and obtain the sample correction coefficient set.

[0042] Further, use the sample weather data set, sample road traffic volume set, sample road surface feature set, sample service life set, and sample correction coefficient set as training data to perform supervised training on the Q correction analysis plugins. During the training process, the model will continuously adjust the network weights and biases, and optimize the network parameters through the backpropagation algorithm until the predicted value output by the network is as close as possible to the actual correction coefficient, so as to achieve convergence and obtain the cycle correction analysis model. Finally, the obtained cycle correction analysis model can output the corresponding cycle correction coefficient according to the input weather data, road traffic volume, road surface features, and service life, and provide optimized cycle correction suggestions for road operation and maintenance.

[0043] Through the cycle correction analysis model based on the BP neural network, dynamic and intelligent road operation and maintenance cycle correction can be realized. Through diverse input data, the model can accurately identify the key factors affecting road damage and adjust the maintenance cycle according to the specific conditions of each road section, thereby improving the scientificity, accuracy, and efficiency of road management.

[0044] Furthermore, the present invention further includes the following steps:

[0045] Use the sample weather data set, sample road traffic volume set, sample road surface feature set, sample service life set, and sample correction coefficient set as training data, and divide the training data into Q equal parts to obtain Q training sets. Use the Q training sets to perform supervised training on the Q correction analysis plugins until convergence to obtain Q converged correction analysis plugins. Integrate and construct the cycle correction analysis model according to the Q converged correction analysis plugins, where the output of the cycle correction analysis model is the mean of the outputs of the Q converged correction analysis plugins.

[0046] Specifically, first, use the sample weather data set, sample road traffic volume set, sample road surface feature set, sample service life set, and sample correction coefficient set as training data, and equally divide the training data into Q parts to obtain Q training sets. This division method is usually used to enhance the generalization ability of the model. By training with different subsets, the risk of overfitting is reduced. Each training set contains a subset of the data and is used to train a specific correction analysis plugin.

[0047] Next, for each subset, use an independent correction analysis plugin for training. Each correction analysis plugin is responsible for learning the corresponding rules and patterns from different training sets, forming Q correction analysis plugin models. Each plugin will learn through training how to output appropriate periodic correction coefficients based on input data such as weather data, traffic volume data, and road surface features. Each correction analysis plugin is trained using the supervised learning method until the error in the training process is minimized and the model output converges, that is, it reaches the convergence state. At this time, each plugin can independently process the input data and output the predicted periodic correction coefficient.

[0048] After the model training is completed, the Q independent correction analysis plugins have converged and can accurately predict the periodic correction coefficients. Although these plugins are independently trained through different data subsets, their goals are the same, that is, to generate periodic correction coefficients based on different input data types (weather, traffic volume, road surface features, etc.). Then, these Q converged correction analysis plugins are integrated to finally obtain a powerful periodic correction analysis model. The output of this model is the mean of the output results of all plugins. By integrating the predictions of multiple plugins, the stability and accuracy of the model can be improved because each plugin may perform better on certain input data, and the integration method can reduce the bias and error of a single plugin.

[0049] Furthermore, the present invention further includes the following steps:

[0050] Obtain a predetermined operation and maintenance cycle; multiply the periodic correction coefficient by the predetermined operation and maintenance cycle, and use the product as the optimized operation and maintenance cycle.

[0051] Specifically, obtain a predetermined operation and maintenance cycle, which refers to the cycle set according to traditional or default maintenance plans without considering specific factors (such as weather, traffic flow, road conditions, etc.). Usually, this cycle is determined based on experience or historical data and may be a fixed time interval (such as once a year or once every six months). Then, multiply the cycle correction coefficient by the predetermined operation and maintenance cycle, and use the product as the optimized operation and maintenance cycle. The optimized operation and maintenance cycle is a corrected operation and maintenance cycle that can automatically adjust the original predetermined operation and maintenance cycle according to the input multi-dimensional data (such as climate, traffic flow, road characteristics, etc.). This cycle is usually more accurate and more in line with actual needs. For example, if a certain road section has higher rainfall and higher traffic flow during a certain period, the system may adjust the maintenance cycle to make the maintenance more frequent during this period, thus effectively avoiding excessive road damage.

[0052] By combining the predetermined operation and maintenance cycle with the cycle correction coefficient, the operation and maintenance cycle of the road can be dynamically adjusted and optimized. This not only improves the efficiency of road maintenance but also better adapts to different environments and road conditions, avoiding resource waste or maintenance lag problems caused by fixed cycles, and realizing more intelligent road maintenance management.

[0053] Step 104: Perform road operation and maintenance inspection in the future time zone according to the optimized operation and maintenance cycle.

[0054] Specifically, finally perform road operation and maintenance inspection in the future time zone according to the optimized operation and maintenance cycle. Through the optimized operation and maintenance cycle, maintenance work can be arranged more accurately in the future time zone, ensuring that the repair work is fully matched with the actual road conditions and avoiding premature or late repair. This process not only improves the efficiency of road maintenance but also minimizes traffic congestion and resource waste, and enhances the long-term operation safety and reliability of the road.

[0055] In summary, the intelligent perception method for road network maintenance provided by the present invention has the following technical effects:

[0056] Obtain historical weather data and road traffic flow data of the target road in the preset historical time zone through Internet of Things monitoring; obtain the road surface characteristics and service life of the target road; based on machine learning, perform operation and maintenance cycle correction analysis according to the historical weather data, road traffic flow data, road surface characteristics and service life to obtain an optimized operation and maintenance cycle; perform road operation and maintenance inspection in the future time zone according to the optimized operation and maintenance cycle; that is to say, by combining historical weather data, road traffic flow data, road surface characteristics and service life, and using a machine learning model to correct and optimize the operation and maintenance cycle, a more accurate and personalized operation and maintenance plan can be realized.

[0057] Embodiment 2. Based on the same inventive concept as the intelligent perception method for road network maintenance in the foregoing embodiment, the present invention also provides an intelligent perception system for road network maintenance. Please refer to the attached Figure 2 , including: an associated information acquisition module 10, configured to obtain historical weather data and road traffic flow data of a target road in a preset historical time period through Internet of Things monitoring; a road surface information acquisition module 20, configured to obtain road surface characteristics and service life of the target road; an operation and maintenance cycle correction analysis module 30, configured to perform operation and maintenance cycle correction analysis based on machine learning according to the historical weather data, road traffic flow data, road surface characteristics and service life to obtain an optimized operation and maintenance cycle; and a road operation and maintenance inspection module 40, configured to perform road operation and maintenance inspection in a future time period according to the optimized operation and maintenance cycle.

[0058] Further, the intelligent perception system for road network maintenance is further configured to: obtain historical weather data of a target road in a preset historical time period through Internet of Things monitoring, where the historical weather data includes average temperature, maximum temperature difference and rainfall; obtain road traffic flow data of the target road in a preset historical time period through Internet of Things monitoring, where the road traffic flow data includes average daily vehicle flow and average daily large vehicle flow of the road.

[0059] Further, the intelligent perception system for road network maintenance is further configured to: obtain road surface characteristics of the target road, where the road surface characteristics include road surface type, road surface material and road surface damage ratio; obtain multiple service durations of multiple sections of the target road, and calculate the average value of the multiple service durations to obtain the service life.

[0060] Further, the intelligent perception system for road network maintenance is further configured to: construct a cycle correction analysis model based on machine learning, and collect sample data to train the cycle correction analysis model until convergence; input the historical weather data, road traffic flow data, road surface characteristics and service life into the cycle correction analysis model for analysis, and output a cycle correction coefficient; perform operation and maintenance cycle correction analysis according to the cycle correction coefficient to obtain an optimized operation and maintenance cycle.

[0061] Furthermore, the intelligent perception system for road network maintenance is also used for: constructing a periodic correction analysis model based on a BP neural network, where the periodic correction analysis model includes Q correction analysis plugins; collecting a sample weather data set, a sample road traffic volume set, a sample road surface feature set, and a sample service life set, and statistically calculating the mean of historical periodic correction coefficients under different sample weather data, sample road traffic volumes, sample road surface features, and sample service lives, denoted as sample correction coefficients, to obtain a sample correction coefficient set; using the sample weather data set, the sample road traffic volume set, the sample road surface feature set, the sample service life set, and the sample correction coefficient set as training data to perform supervised training on the Q correction analysis plugins until convergence, so as to obtain the periodic correction analysis model.

[0062] Furthermore, the intelligent perception system for road network maintenance is also used for: using the sample weather data set, the sample road traffic volume set, the sample road surface feature set, the sample service life set, and the sample correction coefficient set as training data, and equally dividing the training data into Q parts to obtain Q training sets; using the Q training sets to perform supervised training on the Q correction analysis plugins until convergence to obtain Q converged correction analysis plugins; integrating and constructing the periodic correction analysis model according to the Q converged correction analysis plugins, where the output of the periodic correction analysis model is the mean of the outputs of the Q converged correction analysis plugins.

[0063] Furthermore, the intelligent perception system for road network maintenance is also used for: obtaining a predetermined operation and maintenance cycle; multiplying the periodic correction coefficient by the predetermined operation and maintenance cycle, and using the product of the two as the optimized operation and maintenance cycle.

[0064] The various embodiments in this specification are described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The intelligent perception method and specific examples in the first embodiment described above are equally applicable to the intelligent perception system for road network maintenance in this embodiment. Through the detailed description of the intelligent perception method for road network maintenance above, those skilled in the art can clearly know the intelligent perception system for road network maintenance in this embodiment. Therefore, for the sake of simplicity of the specification, it will not be elaborated here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0065] The foregoing description of the disclosed embodiments enables those skilled in the art to practice or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Thus, the invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0066] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention is also intended to include these changes and modifications.

Claims

1. An intelligent perception method for road network maintenance, characterized in that, The method includes: Obtaining historical weather data and road traffic data of a target road in a preset historical time zone through Internet of Things monitoring; Obtaining the road surface characteristics and service life of the target road; Based on machine learning, performing operation and maintenance cycle correction analysis according to the historical weather data, road traffic data, road surface characteristics and service life to obtain an optimized operation and maintenance cycle; Performing road operation and maintenance inspection in a future time zone according to the optimized operation and maintenance cycle.

2. The intelligent perception method for road network maintenance according to claim 1, characterized in that, Obtaining historical weather data and road traffic data of a target road in a preset historical time zone through Internet of Things monitoring, including: Obtaining historical weather data of a target road in a preset historical time zone through Internet of Things monitoring, where the historical weather data includes average temperature, maximum temperature difference and rainfall; Obtaining road traffic data of a target road in a preset historical time zone through Internet of Things monitoring, where the road traffic data includes average daily traffic volume and average daily large vehicle traffic volume.

3. The intelligent perception method for road network maintenance according to claim 2, characterized in that Obtaining the road surface characteristics and service life of the target road, including: Obtaining the road surface characteristics of the target road, where the road surface characteristics include road surface type, road surface material and road surface damage ratio; Obtaining the service durations of multiple sections of the target road, and calculating the mean value of the multiple service durations to obtain the service life.

4. The intelligent perception method for road network maintenance according to claim 3, wherein Based on machine learning, performing operation and maintenance cycle correction analysis according to the historical weather data, road traffic data, road surface characteristics and service life to obtain an optimized operation and maintenance cycle, including: Constructing a cycle correction analysis model based on machine learning, and collecting sample data to train the cycle correction analysis model until convergence; Inputting the historical weather data, road traffic data, road surface characteristics and service life into the cycle correction analysis model for analysis, and outputting a cycle correction coefficient; Performing operation and maintenance cycle correction analysis according to the cycle correction coefficient to obtain an optimized operation and maintenance cycle.

5. The intelligent perception method for road network maintenance according to claim 4, wherein Constructing a cycle correction analysis model based on machine learning, and collecting sample data to train the cycle correction analysis model until convergence, including: Constructing a cycle correction analysis model based on a BP neural network, where the cycle correction analysis model includes Q correction analysis plugins; Collecting a sample weather data set, a sample road traffic set, a sample road surface characteristics set and a sample service life set, and statistically calculating the mean value of historical cycle correction coefficients under different sample weather data, sample road traffic, sample road surface characteristics and sample service life, denoted as sample correction coefficients, to obtain a sample correction coefficient set; Using the sample weather data set, the sample road traffic set, the sample road surface characteristics set, the sample service life set and the sample correction coefficient set as training data to perform supervised training on the Q correction analysis plugins until convergence to obtain the cycle correction analysis model.

6. The intelligent perception method for road network maintenance according to claim 5, wherein Performing supervised training on the Q correction analysis plugins until convergence to obtain the cycle correction analysis model, including: Using the sample weather data set, the sample road traffic set, the sample road surface characteristics set, the sample service life set and the sample correction coefficient set as training data, and equally dividing the training data into Q parts to obtain Q training sets; Supervise and train the Q calibration analysis plugins using the Q training sets until convergence to obtain Q converged calibration analysis plugins; Integrate and construct the periodic calibration analysis model according to the Q converged calibration analysis plugins, where the output of the periodic calibration analysis model is the mean value of the outputs of the Q converged calibration analysis plugins.

7. The intelligent perception method for road network maintenance according to claim 4, characterized in that Perform operation and maintenance period calibration analysis according to the periodic calibration coefficient to obtain an optimized operation and maintenance period, including: Obtain a predetermined operation and maintenance period; Multiply the periodic calibration coefficient by the predetermined operation and maintenance period, and use the product of the two as the optimized operation and maintenance period.

8. The intelligent perception system for road network maintenance is characterized in that, Steps for implementing the intelligent perception method for road network maintenance according to any one of claims 1 to 7, including: An associated information acquisition module, configured to acquire historical weather data and road traffic data of a target road in a preset historical time zone through Internet of Things monitoring; A road surface information acquisition module, configured to acquire the road surface characteristics and service life of the target road; An operation and maintenance period calibration analysis module, configured to perform operation and maintenance period calibration analysis based on machine learning according to the historical weather data, road traffic data, road surface characteristics and service life to obtain an optimized operation and maintenance period; A road operation and maintenance inspection module, configured to perform road operation and maintenance inspection in a future time zone according to the optimized operation and maintenance period.

Images