A method and system for predicting the internal temperature field of cement concrete pavement
By dividing natural regions into three levels and using neural network prediction, combined with meteorological information, the problem of low accuracy in road surface temperature field prediction in existing technologies has been solved, achieving higher prediction accuracy and engineering applicability.
Patent Information
- Application Number
- CN202310411134.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-04-18
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2043-04-18
AI Technical Summary
Existing methods for predicting road surface temperature fields suffer from low accuracy or excessive parameter requirements, making it difficult to meet accuracy and engineering needs.
By dividing the natural areas into three levels and combining meteorological information from meteorological stations, a neural network is used to predict the internal temperature field of cement concrete pavement. The division is further refined using average altitude, river and geological disaster data, and prediction is made by combining meteorological estimates and measured pavement temperature data.
It enables more detailed natural zone division and meteorological estimation, improving the accuracy of predicting the internal temperature field of the road surface and its engineering applicability.
Smart Images

Figure CN116432532B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of environmental monitoring technology, specifically relating to a method and system for predicting the internal temperature field of cement concrete pavement. Background Technology
[0002] The long-term performance of road surfaces, encompassing durability, functionality, and load-bearing capacity, is the result of the combined effects of traffic loads and the environment, with road surface temperature being a crucial environmental factor. Temperature variations directly influence the stress-strain state and material loss evolution of the road surface. Predicting the road surface temperature field is a vital component of research into the long-term performance evolution of road surfaces.
[0003] Currently, research on pavement temperature field prediction methods mainly falls into two categories. One approach utilizes the principles of meteorology and heat conduction, employing numerical analysis methods based on meteorological data, pavement material parameters, and boundary conditions to establish a pavement temperature field prediction model; this is known as the theoretical analysis method. The other approach uses regression analysis based on measured pavement temperature data and meteorological data to establish the relationship between pavement temperature and meteorological factors; this is known as the statistical analysis method. While the theoretical analysis method is highly adaptable, it requires a large number of parameters for modeling and has relatively low accuracy. Although the statistical analysis method suffers from poor regional adaptability in its prediction formulas, it requires fewer parameters to establish the model, is simpler to model, and has higher prediction accuracy, meeting the needs of localized engineering projects. Both existing methods have certain limitations; therefore, a pavement temperature field prediction method with high accuracy and good performance is urgently needed. Summary of the Invention
[0004] This invention aims to address the shortcomings of existing technologies by proposing a method and system for predicting the internal temperature field of cement concrete pavements. It achieves more refined three-level natural zone division by dividing the natural zones into three levels. Simultaneously, it combines meteorological information from weather stations to estimate meteorological conditions at any location within these three levels of natural zones. A neural network is employed, combining meteorological estimates with collected pavement temperature data, to predict the internal temperature field of the pavement.
[0005] To achieve the above objectives, the present invention provides the following solution: a method for predicting the internal temperature field of cement concrete pavement, comprising the following steps:
[0006] S1. Based on average elevation, rivers, and geological hazards, the secondary natural zones are divided into several tertiary natural zones;
[0007] S2. Collect the internal temperature of the road surface to be collected within the three-level natural zone area to obtain internal temperature information;
[0008] S3. Obtain the location information of the road surface to be collected, and obtain the temperature value based on meteorological data;
[0009] S4. Construct a temperature prediction model, and predict the internal temperature field based on the temperature prediction model, the internal temperature information, and the air temperature value to obtain the prediction result.
[0010] Preferably, S1 includes:
[0011] Collect average elevation data, river conditions, and historical geological disaster data within the secondary natural zones;
[0012] The river information includes: whether a river exists, and the area data of the river;
[0013] The geological disaster data includes: whether a geological disaster has occurred, as well as the type, level, and extent of the disaster;
[0014] Thresholds are applied to the average altitude, river area, and geological disaster data, and the secondary natural regions are then divided into tertiary natural regions based on these thresholds, resulting in several tertiary natural regions.
[0015] Preferably, S2 includes:
[0016] The number and depth of test points are determined based on the road surface to be sampled.
[0017] The temperature sensors are deployed based on the number of test points and the burial depth.
[0018] The sampling frequency is set to collect internal temperature information of the road surface, thus obtaining the internal temperature information.
[0019] Preferably, S3 includes:
[0020] Determine the location information of the road surface to be collected, and construct a grid for the third-level natural zone where the road surface to be collected is located using contour line data;
[0021] Determine the distribution of surrounding weather stations and obtain meteorological data from them;
[0022] The temperature at the aforementioned weather station is converted to a virtual sea level based on altitude and the vertical temperature lapse rate.
[0023] The temperature values on the virtual sea level are interpolated and rasterized according to a predetermined grid size;
[0024] The temperature of the virtual sea level raster is interpolated using the Klügün interpolation method to generate a temperature raster map;
[0025] The road surface temperature value is obtained by subtracting the temperature difference due to altitude increase from the grid temperature value on the virtual sea level according to the vertical temperature lapse rate.
[0026] Preferably, the temperature prediction model includes:
[0027] Surface heat flow simulation layer and road surface temperature simulation layer;
[0028] The surface heat flow simulation layer takes the air temperature value, road surface parameters and basic environmental parameters as input and outputs the surface heat flow of the road surface.
[0029] The road surface temperature simulation layer takes the heat flow on the road surface as input and outputs the number of road surface temperature points.
[0030] The present invention also provides a system for predicting the internal temperature field of cement concrete pavement, comprising: a division unit, a data acquisition unit, a meteorological unit, and a construction unit;
[0031] The division unit is used to divide the secondary natural regions based on average altitude, rivers, and geological hazards, resulting in several tertiary natural regions;
[0032] The acquisition unit is used to acquire the internal temperature of the road surface to be acquired within the three-level natural zone area to obtain internal temperature information;
[0033] The meteorological unit is used to acquire the location information of the road surface to be collected and to obtain the temperature value based on the meteorological data;
[0034] The construction unit is used to construct a temperature prediction model, and to predict the internal temperature field based on the temperature prediction model, the internal temperature information and the air temperature value, so as to obtain the prediction result.
[0035] Preferably, the method for obtaining the three-level natural regions by the division unit includes:
[0036] Collect average elevation data, river conditions, and historical geological disaster data within the secondary natural zones;
[0037] The river information includes: whether a river exists, and the area data of the river;
[0038] The geological disaster data includes: whether a geological disaster has occurred, as well as the type, level, and extent of the disaster;
[0039] Thresholds are applied to the average altitude, river area, and geological disaster data, and the secondary natural regions are then divided into tertiary natural regions based on these thresholds, resulting in several tertiary natural regions.
[0040] Preferably, the method by which the acquisition unit obtains the internal temperature information includes:
[0041] The number and depth of test points are determined based on the road surface to be sampled.
[0042] The temperature sensors are deployed based on the number of test points and the burial depth.
[0043] The sampling frequency is set to collect internal temperature information of the road surface, thus obtaining the internal temperature information.
[0044] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0045] This invention discloses a method and system for predicting the internal temperature field of cement concrete pavement. It divides natural zones into three levels, resulting in more refined three-level natural zones. Simultaneously, it combines meteorological information from weather stations to achieve meteorological estimation at any location within these three levels of natural zones. A neural network is employed, combining meteorological estimation with collected pavement temperature data, to predict the internal temperature field of the pavement. Attached Figure Description
[0046] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0047] Figure 1 This is a flowchart illustrating a method for predicting the internal temperature field of a cement concrete pavement according to an embodiment of the present invention.
[0048] Figure 2 This is a schematic diagram of the temperature prediction model structure according to an embodiment of the present invention;
[0049] Figure 3 This is a schematic diagram of the structure of a cement concrete pavement internal temperature field prediction system according to an embodiment of the present invention. Detailed Implementation
[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0051] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0052] Example 1
[0053] like Figure 1 The diagram shown is a flowchart of a method for predicting the internal temperature field of a cement concrete pavement according to this embodiment, including the following steps:
[0054] S1. Based on average elevation, rivers, and geological hazards, the secondary natural zones are divided into several tertiary natural zones;
[0055] The research object of natural regionalization is the natural complex. It studies the characteristics and development patterns of natural complexes, and divides them into regions based on their similarities and differences on the Earth's surface. These regions are then divided into natural units with a hierarchical system of subordinate relationships, expressed as comprehensive natural regions, which can fully reflect the regional differentiation of nature. The national highway natural regionalization is divided into three levels:
[0056] (1) First-level division
[0057] The country is divided into 7 primary regions: I—Northern permafrost region; II—Eastern humid monsoon region; III—Loess Plateau dry-wet transition zone; IV—Southeastern humid and hot region; V—Southwestern humid and warm region; VI—Northwestern arid region; VII—Qinghai-Tibet alpine region.
[0058] (2) Second-level division
[0059] The secondary zoning is based primarily on the humidity coefficient. According to the similarity of highway projects and the differences in surface climate, it is further divided into 33 secondary zones and 19 sub-zones within the 7 primary zoning.
[0060] The humidity coefficient K is the ratio of annual precipitation (mm) to annual evaporation (mm), and is divided into 6 levels according to the size of the K value in the region.
[0061] (3) Three-level division
[0062] The third-level zoning is a further subdivision of the second-level zoning. Each province, municipality, and autonomous region can further subdivide the second-level zoning based on local topography, hydrology, soil quality, and other specific conditions. For details on the names and characteristics of highway natural zoning, please refer to the "Standards for Highway Natural Zoning" (JTJ003-86).
[0063] This embodiment uses natural conditions such as average altitude, rivers, and geological hazards as indicators to divide natural regions into three levels. Specifically,
[0064] The data collected includes average elevation, river conditions, and historical geological disaster data, organized by county, town, or village within the secondary natural region. River conditions include the existence of rivers; if rivers exist, their area data is collected. Geological disaster data includes whether geological disasters have occurred; if so, the disaster type, severity level, and extent are collected.
[0065] Threshold classifications are applied to average altitude, river area, and geological hazard data. In this embodiment, for plains, the average altitude is defined as 350 meters; below 350 meters is considered low altitude, and above 350 meters is considered high altitude. For plateaus, the average altitude is defined as 800 meters; below 800 meters is considered low altitude, and above 800 meters is considered high altitude. The classification of plains and plateaus in this embodiment follows official standards. Regarding river area, this embodiment uses river area and length as indicators to classify rivers into large, medium, and small rivers. Large rivers have a drainage area greater than 5000 km² and a length greater than 200 km; medium rivers have a drainage area greater than 1000 km² but less than 5000 km² and a length greater than 100 km but less than 200 km; and small rivers have a drainage area less than 1000 km² and a length less than 100 km. Geological hazards are classified into four levels—extra-large, large, medium, and small—based on the magnitude of casualties and economic losses.
[0066] Based on the above indicators, the secondary natural regions are divided into tertiary natural regions, resulting in several tertiary natural regions.
[0067] S2. Collect the internal temperature of the road surface to be collected within the third-level natural zone area to obtain internal temperature information;
[0068] Temperature data was collected from the road surface in different third-level natural zones to obtain accurate internal temperature information; specifically...
[0069] First, determine the number and depth of test points based on the road surface to be sampled;
[0070] The number of test points is set according to the actual situation, based on factors such as road length, and the burial depth is set according to actual needs, based on factors such as road thickness. Among these, the road surface, 2cm from the top slab, and the bottom slab are mandatory test points. In this embodiment, the burial depths are set to 5cm, 15cm, 30cm, 50cm, and 80cm, respectively. The burial depth data can be adjusted adaptively according to actual needs; the above burial depth settings are merely illustrative and not intended as limitations.
[0071] Then, the temperature sensors are deployed based on the number of test points and the burial depth;
[0072] Finally, the sampling frequency was set to collect internal temperature information of the road surface to obtain internal temperature information.
[0073] Since the installation of temperature sensors requires operations such as trenching and digging, which damages the original structure of the road surface, a certain waiting period is required after the temperature sensors are installed before formal temperature collection can begin. In this embodiment, the internal temperature information is formally collected after an interval of 7 days.
[0074] S3. Obtain the location information of the road surface to be collected, and obtain the temperature value based on meteorological data;
[0075] In this embodiment, a linear correlation model is obtained using latitude and altitude as parameters.
[0076] First, determine the location information of the road surface to be collected. Then, use contour data to construct a grid for the corresponding third-level natural zone, determine the distribution of surrounding weather stations, and acquire meteorological data from these stations. Finally, convert the existing weather station temperatures to the virtual sea level based on altitude and the vertical temperature lapse rate.
[0077] The temperature values on the virtual sea level are interpolated and rasterized according to a predetermined grid size;
[0078] The temperature raster map is generated by interpolating the virtual sea level raster obtained above using the Klügin interpolation method.
[0079] Finally, the road surface temperature value is obtained by subtracting the temperature difference due to altitude increase from the grid temperature value on the virtual sea level according to the vertical temperature lapse rate.
[0080] S4. Construct a temperature prediction model, and predict the internal temperature field based on the temperature prediction model, internal temperature information, and air temperature value to obtain the prediction results.
[0081] Obtaining the prediction results also requires collecting road surface parameters. In this embodiment, the road surface parameters include: road slab dimensions, strength, and casting time; the basic environmental parameters include: the foundation material beneath the road slab.
[0082] First, establish a linear correlation between meteorological information and internal temperature information.
[0083] A trend analysis was performed on the internal temperature information actually collected in step S2 and the meteorological information obtained in step S3. The trends were basically consistent. Statistical analysis of the internal temperature information and meteorological information showed that there were certain time differences in the linear correlation between the two. Therefore, in this embodiment, the internal temperature information and meteorological data at burial depths of 5cm, 15cm, and 30cm showed basically consistent linear correlation. For the prediction of internal temperature information at 50cm, meteorological data from 15 days ago was used. For the prediction of internal temperature information at 80cm, meteorological data from 28 days ago was used.
[0084] Considering both road surface parameters and basic environmental parameters on internal temperature, this embodiment uses an LSTM neural network as the temperature prediction model, consisting of a forget gate, an input gate, and an output gate. The specific structure is as follows: Figure 2As shown, X represents the input, LSTM1 represents the surface heat flow simulation layer, LSTM2 represents the road surface temperature simulation layer, and YD represents the output. In this embodiment, the temperature prediction model includes a surface heat flow simulation layer and a road surface temperature simulation layer; the surface heat flow simulation layer takes the aforementioned basic environmental parameters, road surface parameters, and the aforementioned air temperature values as input and outputs the surface heat flow of the road surface; the road surface temperature simulation layer takes the output of the surface heat flow simulation layer, namely the surface heat flow of the road surface, as input and outputs the number of road surface temperature points. In this embodiment, the input data of the temperature prediction model is first standardized, and each parameter is controlled within a (0,1) distribution; the temperature prediction model parameters are set as follows: 100,000 training iterations, 22 hours for each training iteration, 2 hours lag time, and 0.001 learning rate.
[0085] Example 2
[0086] like Figure 3 The diagram shown is a structural schematic of a cement concrete pavement internal temperature field prediction system according to this embodiment, including: a division unit, a data acquisition unit, a meteorological unit, and a construction unit;
[0087] The division unit is used to divide the secondary natural regions based on average elevation, rivers, and geological hazards, resulting in several tertiary natural regions;
[0088] The research object of natural regionalization is the natural complex. It studies the characteristics and development patterns of natural complexes, and divides them into regions based on their similarities and differences on the Earth's surface. These regions are then divided into natural units with a hierarchical system of subordinate relationships, expressed as comprehensive natural regions, which can fully reflect the regional differentiation of nature. The national highway natural regionalization is divided into three levels:
[0089] (1) First-level division
[0090] The country is divided into 7 primary regions: I—Northern permafrost region; II—Eastern humid monsoon region; III—Loess Plateau dry-wet transition zone; IV—Southeastern humid and hot region; V—Southwestern humid and warm region; VI—Northwestern arid region; VII—Qinghai-Tibet alpine region.
[0091] (2) Second-level division
[0092] The secondary zoning is based primarily on the humidity coefficient. According to the similarity of highway projects and the differences in surface climate, it is further divided into 33 secondary zones and 19 sub-zones within the 7 primary zoning.
[0093] The humidity coefficient K is the ratio of annual precipitation (mm) to annual evaporation (mm), and is divided into 6 levels according to the size of the K value in the region.
[0094] (3) Three-level division
[0095] The third-level zoning is a further subdivision of the second-level zoning. Each province, municipality, and autonomous region can further subdivide the second-level zoning based on local topography, hydrology, soil quality, and other specific conditions. For details on the names and characteristics of highway natural zoning, please refer to the "Standards for Highway Natural Zoning" (JTJ003-86).
[0096] This embodiment uses natural conditions such as average altitude, rivers, and geological hazards as indicators to divide the natural regions into three levels. Specifically, the method for dividing the natural regions into two levels includes:
[0097] The data collected includes average elevation, river conditions, and historical geological disaster data, organized by county, town, or village within the secondary natural region. River conditions include the existence of rivers; if rivers exist, their area data is collected. Geological disaster data includes whether geological disasters have occurred; if so, the disaster type, severity level, and extent are collected.
[0098] Threshold classifications are applied to average altitude, river area, and geological hazard data. In this embodiment, for plains, the average altitude is defined as 350 meters; below 350 meters is considered low altitude, and above 350 meters is considered high altitude. For plateaus, the average altitude is defined as 800 meters; below 800 meters is considered low altitude, and above 800 meters is considered high altitude. The classification of plains and plateaus in this embodiment follows official standards. Regarding river area, this embodiment uses river area and length as indicators to classify rivers into large, medium, and small rivers. Large rivers have a drainage area greater than 5000 km² and a length greater than 200 km; medium rivers have a drainage area greater than 1000 km² but less than 5000 km² and a length greater than 100 km but less than 200 km; and small rivers have a drainage area less than 1000 km² and a length less than 100 km. Geological hazards are classified into four levels—extra-large, large, medium, and small—based on the magnitude of casualties and economic losses.
[0099] Based on the above indicators, the secondary natural regions are divided into tertiary natural regions, resulting in several tertiary natural regions.
[0100] The data acquisition unit is used to collect the internal temperature of the road surface within the three-level natural zone area to obtain internal temperature information;
[0101] Temperature data is collected from the road surface in different third-level natural zones to obtain real internal temperature information; specifically, the working process of the data collection unit includes:
[0102] First, determine the number and depth of test points based on the road surface to be sampled;
[0103] The number of test points is set according to the actual situation, based on factors such as road length, and the burial depth is set according to actual needs, based on factors such as road thickness. Among these, the road surface, 2cm from the top slab, and the bottom slab are mandatory test points. In this embodiment, the burial depths are set to 5cm, 15cm, 30cm, 50cm, and 80cm, respectively. The burial depth data can be adjusted adaptively according to actual needs; the above burial depth settings are merely illustrative and not intended as limitations.
[0104] Then, the temperature sensors are deployed based on the number of test points and the burial depth;
[0105] Finally, the sampling frequency was set to collect internal temperature information of the road surface to obtain internal temperature information.
[0106] Since the installation of temperature sensors requires operations such as trenching and digging, which damages the original structure of the road surface, a certain waiting period is required after the temperature sensors are installed before formal temperature collection can begin. In this embodiment, the internal temperature information is formally collected after an interval of 7 days.
[0107] The meteorological unit is used to acquire the location information of the road surface to be collected and to obtain the temperature value based on the meteorological data;
[0108] In this embodiment, a linear correlation model is obtained using latitude and altitude as parameters.
[0109] First, determine the location information of the road surface to be collected. Then, use contour data to construct a grid for the corresponding third-level natural zone, determine the distribution of surrounding weather stations, and acquire meteorological data from these stations. Finally, convert the existing weather station temperatures to the virtual sea level based on altitude and the vertical temperature lapse rate.
[0110] The temperature values on the virtual sea level are interpolated and rasterized according to a predetermined grid size;
[0111] The temperature raster map is generated by interpolating the virtual sea level raster obtained above using the Klügin interpolation method.
[0112] Finally, the road surface temperature value is obtained by subtracting the temperature difference due to altitude increase from the grid temperature value on the virtual sea level according to the vertical temperature lapse rate.
[0113] The building unit is used to construct a temperature prediction model and predict the internal temperature field based on the temperature prediction model, internal temperature information, and air temperature values to obtain the prediction results.
[0114] Obtaining the prediction results also requires collecting road surface parameters. In this embodiment, the road surface parameters include: road slab dimensions, strength, and casting time; the basic environmental parameters include: the foundation material beneath the road slab. The specific working process of the construction unit includes:
[0115] First, establish a linear correlation between meteorological information and internal temperature information.
[0116] A trend analysis was performed on the internal temperature information actually collected in step S2 and the meteorological information obtained in step S3. The trends were basically consistent. Statistical analysis of the internal temperature information and meteorological information showed that there were certain time differences in the linear correlation between the two. Therefore, in this embodiment, the internal temperature information and meteorological data at burial depths of 5cm, 15cm, and 30cm showed basically consistent linear correlation. For the prediction of internal temperature information at 50cm, meteorological data from 15 days ago was used. For the prediction of internal temperature information at 80cm, meteorological data from 28 days ago was used.
[0117] Considering both road surface parameters and basic environmental parameters on internal temperature, this embodiment uses an LSTM neural network as the temperature prediction model, consisting of a forget gate, an input gate, and an output gate. The specific structure is as follows: Figure 2 As shown, in this embodiment, the temperature prediction model includes a surface heat flow simulation layer and a road surface temperature simulation layer. The surface heat flow simulation layer takes the aforementioned basic environmental parameters, road surface parameters, and air temperature values as input and outputs the surface heat flow of the road surface. The road surface temperature simulation layer takes the output of the surface heat flow simulation layer (the surface heat flow) as input and outputs the number of road surface temperature points. In this embodiment, the input data of the temperature prediction model is first standardized, and each parameter is controlled within a (0,1) distribution. The temperature prediction model parameters are set as follows: 100,000 training iterations, a single training session length of 22 hours, a lag time of 2 hours, and a learning rate of 0.001.
[0118] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.
Claims
1. A method for predicting the internal temperature field of cement concrete pavement, characterized in that, Includes the following steps: S1. Based on average elevation, rivers, and geological hazards, the secondary natural zones are divided into several tertiary natural zones; S1 includes: collecting average elevation data, river conditions, and historical geological disaster data within the secondary natural zone; the river conditions include: whether rivers exist and their area data; the geological disaster data includes: whether geological disasters have occurred, and the disaster type, level, and extent; thresholding is applied to the average elevation, river area, and geological disaster data, and the secondary natural zone is further divided into tertiary natural zones based on the thresholding, resulting in several tertiary natural zones; S2. Collect the internal temperature of the road surface to be collected within the three-level natural zone area to obtain internal temperature information; S2 includes: determining the number and burial depth of test points based on the road surface to be collected; deploying temperature sensors based on the number of test points and the burial depth; setting the sampling frequency to collect the internal temperature information of the road surface to obtain internal temperature information. S3. Obtain the location information of the road surface to be collected, and obtain the temperature value based on meteorological data; S3 includes: The process involves determining the location of the road surface to be collected, constructing a grid for the third-level natural zone containing the road surface using contour data, identifying the distribution of surrounding meteorological stations, and acquiring meteorological data from these stations. The temperatures from the meteorological stations are then converted to a virtual sea level based on altitude and the vertical temperature lapse rate. The temperature values on the virtual sea level are interpolated and rasterized according to a predetermined grid size. Klüggen interpolation is used to interpolate the temperature of the virtual sea level grid, generating a temperature raster map. Finally, the road surface temperature value is obtained by subtracting the temperature decrease due to altitude from the raster temperature values on the virtual sea level according to the vertical temperature lapse rate. S4. Construct a temperature prediction model, and predict the internal temperature field based on the temperature prediction model, the internal temperature information and the air temperature value to obtain the prediction result; the temperature prediction model includes: a surface heat flow simulation layer and a road surface temperature simulation layer, and uses an LSTM neural network as the temperature prediction model, where LSTM1 represents the surface heat flow simulation layer and LSTM2 represents the road surface temperature simulation layer. The surface heat flow simulation layer takes the air temperature value, road surface parameters, and basic environmental parameters as input and outputs the surface heat flow of the road surface; the road surface temperature simulation layer takes the surface heat flow of the road surface as input and outputs the number of road surface temperature points.
2. A system for predicting the internal temperature field of cement concrete pavement, the system being used to implement the method of claim 1, characterized in that, include: Divide into units, data acquisition units, meteorological units, and construction units; The division unit is used to divide the secondary natural regions based on average altitude, rivers, and geological hazards, resulting in several tertiary natural regions; The acquisition unit is used to acquire the internal temperature of the road surface to be acquired within the three-level natural zone area to obtain internal temperature information; The meteorological unit is used to acquire the location information of the road surface to be collected and to obtain the temperature value based on the meteorological data; The construction unit is used to construct a temperature prediction model, and to predict the internal temperature field based on the temperature prediction model, the internal temperature information and the air temperature value, so as to obtain the prediction result.
3. The system for predicting the internal temperature field of cement concrete pavement according to claim 2, characterized in that, The method for obtaining the three-level natural regions by the division unit includes: Collect average elevation data, river conditions, and historical geological disaster data within the secondary natural zones; The river information includes: whether a river exists, and the area data of the river; The geological disaster data includes: whether a geological disaster has occurred, as well as the type, level, and extent of the disaster; Thresholds are applied to the average altitude, river area, and geological disaster data, and the secondary natural regions are then divided into tertiary natural regions based on these thresholds, resulting in several tertiary natural regions.
4. The system for predicting the internal temperature field of cement concrete pavement according to claim 2, characterized in that, The method by which the acquisition unit obtains the internal temperature information includes: The number and depth of test points are determined based on the road surface to be sampled. The temperature sensors are deployed based on the number of test points and the burial depth. The sampling frequency is set to collect internal temperature information of the road surface, thus obtaining the internal temperature information.
Citation Information
Patent Citations
Mass concrete temperature monitoring data prediction method and system and medium
CN114742260A
