Pavement temperature monitoring method, device and equipment based on road meteorological analysis

Abstract

The present invention provides a pavement temperature monitoring method, device, and equipment based on road meteorological analysis, comprising: step 1, obtaining material property information, measured meteorological data, and measured pavement temperature values ​​corresponding to the road surface to be monitored to construct a training data set for a neural network model; step 2, correcting the initial pavement temperature; step 3, determining the pavement temperature of the road surface to be monitored after heating and cooling based on the corrected initial pavement temperature and the training data set using a multi-factor heat conversion formula configured within the neural network model; step 4, correcting the model parameters of the training data set and the neural network model based on the heated and cooled pavement temperature and the measured pavement temperature values; and step 5, repeating steps 1 to 4 until the neural network model training is completed. The present invention not only significantly improves the stability and accuracy of pavement temperature monitoring, but also effectively reduces the cost and technical requirements required for pavement temperature monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of road surface temperature monitoring, and in particular to a road surface temperature monitoring method, device and equipment based on road meteorological analysis. Background Art

[0002] As abnormal road conditions have an increasingly significant impact on travel, current methods for road condition monitoring typically include fixed equipment monitoring, mobile vehicle-mounted monitoring, and remote sensing monitoring. Mobile vehicle-mounted monitoring and fixed equipment monitoring have the ability to monitor road slipperiness and road conditions; fixed equipment monitoring typically only monitors road slipperiness and road conditions within a specified range; mobile vehicle-mounted monitoring, compared to fixed equipment monitoring, adds temperature monitoring capabilities. The operation of on-board equipment requires personnel to drive the vehicle to collect basic data, and unexpected equipment damage and spare parts replacement can lead to unstable monitoring status. Remote sensing monitoring is the most comprehensive, but it requires significant initial funding, high technical requirements, and low demand. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a pavement temperature monitoring method, device and equipment based on road meteorological analysis, which can not only significantly improve the stability and accuracy of pavement temperature monitoring, but also effectively reduce the cost and technical requirements required for pavement temperature monitoring.

[0004] In a first aspect, the present invention provides a road surface temperature monitoring method based on road meteorological analysis, comprising:

[0005] Step 1: Obtain material property information, measured meteorological data, and measured road surface temperature values ​​corresponding to the road surface to be monitored to construct a training data set for a neural network model; wherein the custom layer of the neural network model is configured with a multi-factor heat conversion formula;

[0006] Step 2: Correcting the initial road surface temperature corresponding to the road surface to be monitored;

[0007] Step 3: Using the multi-factor heat conversion formula configured in the neural network model, based on the corrected initial road surface temperature and the training data set, determine the temperature change pattern corresponding to the road surface to be monitored and the corresponding road surface temperature after heating and cooling;

[0008] Step 4: calibrating the training data set and the model parameters of the neural network model based on the road surface temperature after heating and cooling and the measured road surface temperature;

[0009] Step 5: Based on the corrected training data set and the corrected neural network model, repeat steps 1 to 4 until the neural network model training is completed. The trained neural network model is used to monitor the temperature of the road surface to be monitored based on the material property information and meteorological time series data corresponding to the road surface to be monitored.

[0010] In one embodiment, correcting the initial road surface temperature corresponding to the road surface to be monitored includes:

[0011] If the current iteration process is the first iteration process, the initial road surface temperature corresponding to the road surface to be monitored is corrected based on the measured road surface temperature value;

[0012] If the current iteration process is not the first iteration process, the initial road surface temperature corresponding to the road surface to be monitored is corrected based on the current road surface temperature of the road surface to be monitored at a specified time on the previous day output by the neural network model.

[0013] In one embodiment, a multi-factor heat conversion formula configured within a neural network model is used to determine the temperature change pattern corresponding to the road surface to be monitored and the corresponding road surface temperature after heating and cooling, based on the corrected initial road surface temperature and a training data set, including:

[0014] Determine the absolute absorbed energy value based on the corrected initial road surface temperature and the training data set using a multi-factor heat conversion formula configured within the neural network model;

[0015] Based on the positive and negative values ​​of the absolute absorbed energy values, a temperature change mode corresponding to the road surface to be monitored is determined, where the temperature change mode includes a heating mode or a cooling mode;

[0016] According to the temperature change pattern and the unit temperature rise representative value, the road surface temperature after heating and cooling corresponding to the road surface to be monitored is determined.

[0017] In one embodiment, the measured meteorological data includes wind speed data and air temperature data; the multi-factor heat conversion formula includes a heat conduction energy conversion formula, a heat radiation energy conversion formula, and a convection heat transfer energy conversion formula;

[0018] The expression of heat conduction energy conversion formula is:

[0019] ;

[0020] in, is the energy value generated during heat conduction; For the moment; is the unit radiation road area; is the thermal conductivity of the material; For the road surface to be monitored at time Air temperature data below; For the road surface to be monitored at time Current road surface temperature under is the coordinate of the heat transfer direction;

[0021] The expression of thermal radiation energy conversion formula is:

[0022] ;

[0023] in, is the energy value generated during the thermal radiation process; For the moment; is the unit radiation road area; is solar radiation; For the Thermal characteristic wavelength absorptivity of each thermal characteristic wave; For the The proportion of thermal characteristic wave energy of each thermal characteristic wave;

[0024] The expression of the convective heat transfer energy conversion formula is:

[0025]

[0026] in, is the energy value generated during the convection heat transfer process; is the specific heat capacity of the material; For quality; For the road surface to be monitored at time The road surface temperature changes below; is the corrected initial road surface temperature; is the road surface temperature when convective heat transfer begins; For the road surface to be monitored at time Air temperature data below; is the standard deviation; is the mean; is the wind speed data; For the moment Wind speed data to time The duration when the wind speed value remains unchanged between the wind speed data.

[0027] In one embodiment, based on the road surface temperature after heating and cooling and the measured road surface temperature, the training data set and the model parameters of the neural network model are calibrated, including:

[0028] For any parameter to be corrected in the training data set and the model parameters of the neural network model, the parameter is corrected based on the road surface temperature after heating and cooling and the measured road surface temperature value, and the other parameters except the parameter are fixed.

[0029] In one embodiment, the parameters to be calibrated include the specific heat capacity of the material in the training data set and the unit temperature rise representative value of the neural network model.

[0030] In a second aspect, the present invention further provides a road surface temperature monitoring device based on road meteorological analysis, comprising:

[0031] A dataset construction module is used to obtain material property information, measured meteorological data, and measured road surface temperature values ​​corresponding to the road surface to be monitored, in order to construct a training dataset for the neural network model; wherein the custom layer of the neural network model is configured with a multi-factor heat conversion formula;

[0032] An initial temperature correction module, used to correct the initial road surface temperature corresponding to the road surface to be monitored;

[0033] The post-change temperature prediction module is used to determine the temperature change pattern of the road surface to be monitored and the corresponding road surface temperature after heating and cooling based on the corrected initial road surface temperature and the training data set using the multi-factor heat conversion formula configured in the neural network model;

[0034] A parameter correction module is used to correct the model parameters of the training data set and the neural network model based on the road surface temperature after heating and cooling and the measured road surface temperature value;

[0035] The repeated execution module is used to repeat the data set construction module, the initial temperature correction module, the changed temperature prediction module, and the parameter correction module based on the corrected training data set and the corrected neural network model until the neural network model training is completed. The trained neural network model is used to monitor the temperature of the road surface to be monitored based on the material property information corresponding to the road surface to be monitored and the meteorological time series data.

[0036] In a third aspect, the present invention further provides an electronic device comprising a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement any one of the methods provided in the first aspect.

[0037] In a fourth aspect, the present invention further provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement any one of the methods provided in the first aspect.

[0038] The present invention provides a pavement temperature monitoring method, device and equipment based on road meteorological analysis, including: step 1, obtaining material property information, measured meteorological data and measured pavement temperature values ​​corresponding to the road surface to be monitored to construct a training data set of a neural network model; wherein the custom layer of the neural network model is configured with a multi-factor heat conversion formula; step 2, correcting the initial pavement temperature corresponding to the road surface to be monitored; step 3, determining the temperature change pattern corresponding to the road surface to be monitored and its corresponding pavement temperature after heating and cooling based on the multi-factor heat conversion formula configured in the neural network model and the corrected initial pavement temperature and the training data set; step 4, correcting the model parameters of the training data set and the neural network model based on the pavement temperature after heating and cooling and the measured pavement temperature values; step 5, repeating steps 1 to 4 based on the corrected training data set and the corrected neural network model until the neural network model training is completed, and the trained neural network model is used to monitor the temperature of the road surface to be monitored based on the material property information and meteorological time series data corresponding to the road surface to be monitored. The above method focuses on the obvious and hidden changes in the temperature of road materials under the influence of meteorological factors. Combined with a time series-based neural network model, it conducts direct material property analysis of road pavement temperature and monitors and predicts pavement temperature at the data level. This can not only significantly improve the stability and accuracy of pavement temperature monitoring, but also effectively reduce the cost and technical requirements required for pavement temperature monitoring.

[0039] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or understood by practicing the present invention. The purposes and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0040] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 A schematic flow chart of a road surface temperature monitoring method based on road meteorological analysis provided by an embodiment of the present invention;

[0043] Figure 2A technical framework diagram of a road surface temperature monitoring method based on road meteorological analysis provided by an embodiment of the present invention;

[0044] Figure 3 A schematic structural diagram of a road surface temperature monitoring device based on road meteorological analysis provided by an embodiment of the present invention;

[0045] Figure 4 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0047] Currently, fixed equipment monitoring has a limited range, mobile monitoring using vehicle-mounted equipment suffers from poor stability, and remote sensing monitoring places high demands on funding and technical expertise. Therefore, the present invention provides a method, device, and equipment for monitoring pavement temperature based on road meteorological analysis. These methods significantly improve the stability and accuracy of pavement temperature monitoring while also effectively reducing the cost and technical requirements required.

[0048] To facilitate understanding of this embodiment, a road surface temperature monitoring method based on road meteorological analysis disclosed in an embodiment of the present invention is first described in detail. Figure 1 The flowchart of a road surface temperature monitoring method based on road meteorological analysis is shown, and the method mainly includes the following steps 1 to 5:

[0049] Step 1: Obtain material property information, measured meteorological data, and measured road surface temperature values ​​corresponding to the road surface to be monitored to construct a training data set for the neural network model.

[0050] The neural network model's custom layer configuration includes multi-factor heat conversion formulas, including heat conduction energy conversion formulas, thermal radiation energy conversion formulas, and convection heat transfer energy conversion formulas. Material property information includes thermal characteristic wavelength absorptivity, material thermal conductivity, and material specific heat capacity. The measured meteorological data includes wind speed and air temperature data corresponding to the road surface to be monitored. Specifically, the training dataset includes thermal characteristic wavelength absorptivity, material thermal conductivity, material specific heat capacity, wind speed data, air temperature data, and measured road surface temperature values. These data serve as model inputs, while measured road surface temperature values ​​serve as training labels.

[0051] Step 2: Correct the initial road surface temperature corresponding to the road surface to be monitored.

[0052] In one example, when the model is trained for the first time, the initial road surface temperature can be derived based on the measured road surface temperature value, and the derived initial road surface temperature can be used to replace the original initial road surface temperature; in another example, when the model is not trained for the first time, the original initial road surface temperature can be replaced based on the current road surface temperature at a specified time the previous day output by the neural network.

[0053] Step 3: Determine the temperature change pattern of the road surface to be monitored and its corresponding road surface temperature after heating and cooling based on the corrected initial road surface temperature and the training data set through the multi-factor heat conversion formula configured in the neural network model.

[0054] In one example, the corrected initial road surface temperature and the thermal characteristic wavelength absorptivity, material thermal conductivity, material specific heat capacity, wind speed data, and air temperature data included in the training dataset are input into the heat conduction energy conversion formula, the thermal radiation energy conversion formula, and the convection heat transfer energy conversion formula, respectively, to obtain the absolute absorbed energy value. Based on the positive or negative sign of the absolute absorbed energy value, the corresponding road surface temperature after heating or cooling is calculated. The road surface temperature after heating or cooling can also be referred to as the road surface temperature after heating or cooling.

[0055] Step 4: Based on the road surface temperature after heating and cooling and the measured road surface temperature value, the training data set and the model parameters of the neural network model are calibrated.

[0056] The parameters to be corrected in the training dataset and the neural network model parameters include the material specific heat capacity in the training dataset and the representative value per unit temperature rise of the neural network model. In one example, the road surface temperature after temperature adjustment can be compared with the measured road surface temperature, and the material specific heat capacity and the representative value per unit temperature rise can be corrected based on the comparison results. Alternatively, the correction process can be as follows: the value of one parameter is corrected while the value of the other parameter is fixed.

[0057] Step 5: Based on the corrected training data set and the corrected neural network model, repeat steps 1 to 4 until the neural network model training is completed. The trained neural network model is used to monitor the temperature of the road surface to be monitored based on the material property information and meteorological time series data corresponding to the road surface to be monitored.

[0058] The pavement temperature monitoring method based on road meteorological analysis provided by the embodiment of the present invention focuses on the obvious and hidden changes in the temperature of road materials under the influence of meteorological factors, combines a time series-based neural network model to perform direct material property analysis of road pavement temperature, and monitors and predicts pavement temperature at the data level. This can not only significantly improve the stability and accuracy of pavement temperature monitoring, but also effectively reduce the cost and technical requirements required for pavement temperature monitoring.

[0059] For ease of understanding, the present invention first explains the principle of the road surface temperature monitoring method based on road meteorological analysis:

[0060] Concrete and asphalt are the most commonly used road materials. Their inherent characteristics will cause them to absorb and release heat under the influence of weather, which will affect the actual temperature of the road, specifically manifesting as road icing and overheating. Taking asphalt as an example, the specific heat capacity of pavements paved with the most common asphalt mixture can reach 1.68 kJ / kg.°C and a thermal conductivity of 3.05 W / (m*K); in contrast, the specific heat capacity of concrete pavements is approximately 0.97 kJ / kg.°C, with a thermal conductivity of λ = 1.51 W / (m*K). During road use, the road surface layer is most susceptible to external influences. The energy that can affect this surface layer is primarily of two types: atmospheric temperature, and heat exchange between the remaining road layers. In this embodiment of the present invention, due to temperature magnitude and variation, the actual pavement surface height is ignored for ease of calculation, with each surface layer considered a horizontal plane. During the study, concrete pavements exhibited higher thermal conductivity and reflectivity than asphalt pavements, and were less sensitive to temperature. This results in higher heating and cooling rates and higher extreme temperatures for the remaining layers of asphalt pavements under the same conditions than for concrete pavements. Furthermore, thermal spectrum mapping technology is commonly used on asphalt roads, such as highways, so the impact of concrete pavements is less than that of asphalt pavements under the same conditions.

[0061] In the conversion relationship between energy and temperature, we know that , To absorb or release heat, is the specific heat capacity of the material, For material quality, = is the temperature change. 1 cubic meter of concrete mixture weighs approximately 2400 kg, and the energy required to raise the overall temperature by 1°C is approximately 2328 kJ, with the energy required per unit surface area being 93.12 kJ. 1 cubic meter of asphalt mixture weighs approximately 2486 kg, and when converted to a 25 square meter surface layer, the energy required to raise the overall temperature by 1°C is approximately 4176.48 kJ, with the energy required per unit surface area being 167.06 kJ. Heat transfer between gases and solids is determined according to Fourier's law:

[0062] ;

[0063] is the heat flux density, which is the heat transfer rate per unit area of ​​the road surface in the x direction perpendicular to the road surface, and the unit is ; T is the air temperature data Current road surface temperature The difference between is the coordinate of the heat transfer direction; is the thermal conductivity of the material. The process of heat transfer from the air temperature to the road surface, taking the hot summer climate as an example, is divided into three stages: Stage 1: There is a positive temperature gradient difference between the atmosphere and the road surface. At this time, the heat transfer rate is a gently rising curve, and the heat transfer direction of the heat conduction effect is: atmosphere-road surface; Stage 2: As heat transfer progresses, the temperature gradient difference between the atmosphere and the road surface approaches. At this time, the work done by heat conduction and convection heat transfer decreases, the heat transfer rate value tends to 0, and thermal radiation becomes the main factor of change. The heat transfer direction of the heat conduction effect is: atmosphere-road surface; Stage 3: Due to the heat absorption and release properties of the road surface material, the positive heat conduction effect caused by the temperature gradient difference has turned into a negative heat conduction effect. Thermal radiation becomes the main factor in the temperature increase of the road surface material. At this time, the heat transfer direction of the heat conduction effect is: road surface-air. During the stage-by-stage process, the main contributing factors are thermal radiation absorption, while the main directional factors are thermal conduction and convective heat transfer. The state of convective heat transfer varies depending on the wind speed and intensity of the wind. At a certain intensity of convective heat transfer, when the difference between air temperature and road surface temperature reaches a threshold, the rising and falling temperature factors cancel each other out, the road surface temperature stops changing, and remains relatively static. This threshold is the static threshold, and the stage is the termination stage. At this stage, the state of the stage is that thermal conduction and convective heat transfer offset the thermal radiation effect. This threshold is used as the model calibration point, and the difference between the road surface temperature at this threshold and the measured road surface temperature is used as the calibration value. This calibration value is eliminated by changing the physical characteristics of the material. Using the radiation-temperature and energy-temperature relationships, wind speed and air temperature are combined with the physical characteristics of the material to calculate the absolute energy absorbed per unit area of ​​the road surface and derive the road surface temperature. A time series model, combined with the seasonal characteristics of wind speed and temperature, predicts future wind speed and air temperature, and derives the future invisible changes in road surface temperature for the purpose of thermal spectrum mapping.

[0064] Based on the above principles, the present invention provides a specific implementation of a road surface temperature monitoring method based on road meteorological analysis. Figure 2The technical framework diagram of a road surface temperature monitoring method based on road meteorological analysis is shown, which includes a model training phase and a model application phase. The model training phase includes the following steps: presetting material property information and collecting measured meteorological data; calculating thermal radiation and heat conduction based on the material property information, and calculating convective heat transfer based on the measured meteorological data; calculating the absolute absorbed energy based on thermal radiation, heat conduction, and convective heat transfer using an energy conversion formula; calculating the road surface temperature after heating and cooling from a unit temperature rise representative value using explicit change calculation; collecting the initial road surface temperature, performing a difference correction based on the road surface temperature after heating and cooling, and recalculating the corrected heat conduction; repeating this process until the neural network training is completed. The model application phase includes the following steps: presetting material property information and collecting meteorological time series data; calculating thermal radiation and heat conduction based on the material property information, calculating convective heat transfer based on the meteorological time series data, and predicting the wind speed data and air temperature data of the road surface to be monitored at the monitoring time; through the energy conversion formula, based on thermal radiation, heat conduction and convective heat transfer, combined with the implicit changes in the wind speed data and air temperature data of the road surface to be monitored at the monitoring time, the current road surface temperature of the road surface to be monitored at the monitoring time is determined.

[0065] The specific implementation process is as follows:

[0066] For the aforementioned step S102, an embodiment of the present invention provides an implementation method for constructing a training data set for a neural network model, including: using wind speed data and air temperature data in the measured meteorological data, as well as thermal characteristic wavelength absorptivity, material thermal conductivity and material specific heat capacity contained in the material property information, as model inputs, and using the measured road surface temperature value as a training label to obtain a training data set for the neural network model.

[0067] Regarding the aforementioned step S104, the embodiment of the present invention provides an implementation method for correcting the initial road surface temperature corresponding to the road surface to be monitored, which mainly includes the following two scenarios:

[0068] Scenario 1: If the current iteration is the first, the initial road surface temperature corresponding to the monitored road surface is corrected based on the measured road surface temperature. For example, if the original initial road surface temperature is set to be the same as the air temperature, the initial road surface temperature can be corrected by directly using the actual road surface temperature value collected by the sensor. This actual road surface temperature value can be used to replace the original initial road surface temperature value.

[0069] Scenario 2: If the current iteration is not the first, the initial road surface temperature corresponding to the monitored road surface is corrected based on the current road surface temperature at a specified time the previous day, as output by the neural network model. For example, when subsequently correcting the initial road surface temperature, the final temperature value of the previous day (i.e., the current road surface temperature at midnight, as output by the neural network model) can be used to replace the original initial road surface temperature value.

[0070] Among the three major factors that generate heat exchange, thermal radiation is an absolute positive factor, while convection heat transfer and heat conduction are directional factors. Before explaining the aforementioned step S106, the embodiment of the present invention explains the three processes of thermal radiation, heat conduction, and convection heat transfer respectively.

[0071] Thermal radiation: Thermal radiation can be calculated using the following formula:

[0072] ;

[0073] ;

[0074] in, is the effective radiation, the unit is W / m2; is the absorption coefficient, including the thermal characteristic wave energy occupancy ratio and asphalt mixture absorption rate; is solar radiation; is the proportion of thermal characteristic wave energy; is the absorptivity of each thermal characteristic wavelength.

[0075] Table 1 Wavelength absorption table

[0076]

[0077] Since the energy occupancy ratio of thermal characteristic waves is related to the absorption rate of asphalt mixture and the temperature of the radiation source, and the temperature of the radiation source is the air temperature, slight changes will be made according to the temperature of the radiation source during model calculation, and the main change bandwidth is between 750-1400 and 1400-3000.

[0078] Convective heat transfer: The temperature change state in convection heat transfer conforms to the normal distribution:

[0079] ;

[0080] ;

[0081] ;

[0082] ;

[0083] ;

[0084] Where F(t) is the current road surface temperature; f(t) is the temperature change rate; = , is the standard deviation; T is the road surface temperature when convective heat transfer starts; v is the current wind speed. When v>0.87m / s, convective heat transfer is considered to have started. =0( is the mean); is the change under the influence of wind speed; is the current air temperature; is the initial road surface temperature of the day; for The wind speed at the time The duration when the wind speed value remains unchanged between the wind speeds at different moments, in minutes.

[0085] Heat conduction: Heat conduction can be calculated using the following formula:

[0086] ;

[0087] is the heat flux density, which is the heat transfer rate per unit area of ​​the road surface in the x direction perpendicular to the road surface, and the unit is ; x is the coordinate of the heat transfer direction; k is the thermal conductivity of the material; For the road surface to be monitored at time Air temperature data below; For the road surface to be monitored at time The current road surface temperature.

[0088] In addition, the relationship between radiation and temperature is:

[0089] ;

[0090] ;

[0091] Among them, S is the unit radiation road area, the unit is , t is time in seconds. When calculating M, the principle of uniform distribution of heat energy in the surface layer of 0.4m is followed, and the initial value of x in the direction perpendicular to the road surface is set to 1.

[0092] On this basis, for the aforementioned step S106, an embodiment of the present invention provides an implementation method for determining the temperature change pattern corresponding to the road surface to be monitored and its corresponding road surface temperature after heating and cooling based on the corrected initial road surface temperature and the training data set through a multi-factor heat conversion formula configured in the neural network model, including the following steps a to d.

[0093] Step a, determine whether the model calibration point is reached, and execute the subsequent steps when the model calibration point is reached. In one example, under normal temperature conditions, in the heat exchange generated by the three most basic factors, before the threshold value is deduced, and The heat conduction direction is determined, and then the stage process is distinguished by the wind speed of 0.87m / s at the beginning of convection heat transfer. When v<0.87, the stage process is considered incomplete and there is no threshold; when v>0.87, the stage process is considered incomplete and there is no threshold; The obtained threshold is the judgment condition. When it is satisfied Under this condition, is the stationary threshold, that is, the model correction point is reached.

[0094] Step b: Determine the absolute absorbed energy value based on the corrected initial road surface temperature and the training data set using a multi-factor heat conversion formula configured within the neural network model.

[0095] In one example, the radiation-energy relationship is derived from the energy-temperature relationship and the radiation-temperature relationship. The embodiments of the present invention provide heat conduction energy conversion formulas, heat radiation energy conversion formulas, and convection heat transfer energy conversion formulas. Specifically:

[0096] The expression of heat conduction energy conversion formula is:

[0097] ;

[0098] in, is the energy value generated during heat conduction; For the moment; is the unit radiation road area; is the thermal conductivity of the material; For the road surface to be monitored at time Air temperature data below; For the road surface to be monitored at time Current road surface temperature under is the coordinate of the heat transfer direction;

[0099] The expression of thermal radiation energy conversion formula is:

[0100] ;

[0101] in, is the energy value generated during the thermal radiation process; For the moment; is the unit radiation road area; is solar radiation; For the Thermal characteristic wavelength absorptivity of each thermal characteristic wave; For the The proportion of thermal characteristic wave energy of each thermal characteristic wave;

[0102] The expression of the convective heat transfer energy conversion formula is:

[0103]

[0104] in, is the energy value generated during the convection heat transfer process; is the specific heat capacity of the material; For quality; For the road surface to be monitored at time The road surface temperature changes below; is the corrected initial road surface temperature; is the road surface temperature when convective heat transfer begins; For the road surface to be monitored at time Air temperature data below; is the standard deviation; is the mean; is the wind speed data; For the moment Wind speed data to time The duration when the wind speed value remains unchanged between the wind speed data.

[0105] According to the above formula, the heat generated by the three most basic factors of thermal radiation, heat conduction and convection heat transfer can be calculated respectively. For example, the vector sum of the heat generated by the above three most basic factors can be used as the absolute absorbed energy value.

[0106] Step c: determining a temperature change mode corresponding to the road surface to be monitored based on the positive or negative value of the absolute absorbed energy value, where the temperature change mode includes a heating mode or a cooling mode.

[0107] In one example, before convective heat transfer begins, heat conduction and thermal radiation act on the road surface. The start time is 00:15 in the morning of the same day, and the direction is atmosphere-road. When the absolute absorbed energy value is negative, the road radiates energy to the atmosphere, and this is the cooling mode; when the absolute absorbed energy value is positive, the atmosphere radiates energy to the road, and this is the warming mode.

[0108] Step d: determining the road surface temperature after heating and cooling corresponding to the road surface to be monitored according to the temperature change pattern and the unit temperature rise representative value, where the road surface temperature after heating and cooling includes the road surface temperature after heating or the road surface temperature after cooling.

[0109] The unit temperature rise representative value is the energy required per unit area of ​​the surface layer, which is 167.06 kJ. Based on the above temperature change pattern and the unit temperature rise representative value, the temperature increase / decrease calculation is performed to deduce the road surface temperature after the temperature increase / decrease.

[0110] For the aforementioned step S108, an embodiment of the present invention provides an implementation method for correcting the model parameters of the training data set and the neural network model based on the road surface temperature after heating and cooling and the actual road surface temperature value. For any parameter to be corrected in the training data set and the model parameters of the neural network model, the parameter is corrected based on the road surface temperature after heating and cooling and the actual road surface temperature value, and other parameters except this parameter are fixed.

[0111] In one example, the correction process of the unit temperature rise representative value is as follows: the road temperature after heating / cooling is compared with the measured road surface temperature value, and the unit temperature rise representative value is corrected by the neural network model.

[0112] In one example, the correction process for material specific heat capacity is as follows: Due to the uncertainty of human factors, the fixed values ​​of material properties in the model need to be corrected. The correction values ​​include: characteristic wavelength absorptivity, material specific heat capacity, and thermal conductivity. In this embodiment of the present invention, the primary research object is the explicit and implicit temperature changes of asphalt pavement under meteorological conditions. The variation span is relatively small, so the correction for characteristic wavelength absorptivity is negligible. The thermal conduction phenomenon and the representative value of unit area temperature rise are affected by thermal conductivity and specific heat capacity, respectively. The heat transfer direction of thermal conduction can be determined by the difference between air temperature and pavement temperature. If the difference is positive, heat transfer occurs from air to pavement; if it is negative, heat transfer occurs from pavement to air. The current pavement temperature calculated by the neural network model is compared with the measured pavement temperature. The difference is the correction factor for the errors in specific heat capacity and thermal conductivity. This correction value is numerically related to the specific heat capacity. The representative value of unit area temperature rise is derived from the specific heat capacity. Therefore, the first correction factor should be the material specific heat capacity. When the correction value is known, the ratio of specific heat capacity to thermal conductivity is a constant. The correction value can be eliminated by numerically changing the numerator and denominator of the ratio. The change in the numerator and denominator cannot exceed 5% of the original value.

[0113] Furthermore, during the calibration process, the parameter increment or decrement can be dynamically adjusted. For example, if a small increase in the unit temperature rise representative value does not significantly change the current road surface temperature output by the neural network, the increment of the unit temperature rise representative value can be appropriately increased.

[0114] After the neural network model is trained, it can be used in actual pavement temperature monitoring tasks. Specifically, the following steps are performed: First, the material property information and meteorological time series data corresponding to the road surface to be monitored are obtained; then, based on the meteorological time series data, the wind speed and air temperature data for the road surface to be monitored are predicted; finally, the material property information, wind speed data for the road surface to be monitored, and air temperature data for the road surface to be monitored at the time of monitoring are input into the trained neural network model to monitor the temperature of the road surface to be monitored.

[0115] In summary, the model inputs are characteristic wavelength absorptivity, material thermal conductivity, material specific heat capacity, wind speed, and air temperature. Wind speed and air temperature are temporal factors, and wind speed variations are disordered but exhibit distinct seasonal characteristics. Time series analysis can be used to determine wind speed and air temperature. Since the ultimate goal of the present embodiment is to determine the explicit and implicit changes in pavement temperature, specifically the changes and trends in pavement temperature under the influence of external factors, seasonality is a necessary element in this sequence analysis and is retained. Time series analysis is used to predict air temperature and wind speed, and the predicted values ​​and the material's physical characteristics are incorporated into the formula to calculate the future implicit changes in pavement temperature. The current explicit change in pavement temperature is calculated by incorporating the current wind speed and temperature and the material's physical characteristics into the formula. This embodiment of the present invention focuses on the explicit and implicit changes in pavement material temperature under the influence of meteorological factors. By combining this with a time series-based neural network model, direct material property analysis is performed on road pavement temperature, enabling monitoring and prediction of pavement temperature at the data level. This significantly improves the stability and accuracy of pavement temperature monitoring while also effectively reducing the cost and technical requirements required for monitoring pavement temperature.

[0116] Based on the above embodiments, the present invention provides a road surface temperature monitoring device based on road meteorological analysis. Figure 3 The schematic diagram of a road surface temperature monitoring device based on road meteorological analysis is shown. The device mainly includes the following parts:

[0117] The data set construction module 302 is used to obtain material property information, measured meteorological data, and measured road surface temperature values ​​corresponding to the road surface to be monitored, so as to construct a training data set for the neural network model; wherein the custom layer of the neural network model is configured with a multi-factor heat conversion formula;

[0118] An initial temperature correction module 304 is used to correct the initial road surface temperature corresponding to the road surface to be monitored;

[0119] The post-change temperature prediction module 306 is configured to determine the temperature change pattern corresponding to the road surface to be monitored and the corresponding road surface temperature after heating and cooling based on the corrected initial road surface temperature and the training data set using the multi-factor heat conversion formula configured in the neural network model;

[0120] A parameter correction module 308 is used to correct the model parameters of the training data set and the neural network model based on the road surface temperature after heating and cooling and the measured road surface temperature value;

[0121] The repeated execution module 310 is used to repeat the data set construction module, the initial temperature correction module, the changed temperature prediction module, and the parameter correction module based on the corrected training data set and the corrected neural network model until the neural network model training is completed. The trained neural network model is used to monitor the temperature of the road surface to be monitored based on the material property information and meteorological time series data corresponding to the road surface to be monitored.

[0122] The pavement temperature monitoring device based on road meteorological analysis provided by the embodiment of the present invention focuses on the obvious and hidden changes in the temperature of road materials under the influence of meteorological factors, combines a neural network model based on time series, conducts direct material property analysis of road pavement temperature, and monitors and predicts pavement temperature at the data level. This can not only significantly improve the stability and accuracy of pavement temperature monitoring, but also effectively reduce the cost and technical requirements required for pavement temperature monitoring.

[0123] In one embodiment, the initial temperature correction module 304 is specifically configured to:

[0124] If the current iteration process is the first iteration process, the initial road surface temperature corresponding to the road surface to be monitored is corrected based on the measured road surface temperature value;

[0125] If the current iteration process is not the first iteration process, the initial road surface temperature corresponding to the road surface to be monitored is corrected based on the current road surface temperature of the road surface to be monitored at a specified time on the previous day output by the neural network model.

[0126] In one embodiment, the post-change temperature prediction module 306 is specifically configured to:

[0127] Determine the absolute absorbed energy value based on the corrected initial road surface temperature and the training data set using a multi-factor heat conversion formula configured within the neural network model;

[0128] Based on the positive and negative values ​​of the absolute absorbed energy values, a temperature change mode corresponding to the road surface to be monitored is determined, where the temperature change mode includes a heating mode or a cooling mode;

[0129] According to the temperature change pattern and the unit temperature rise representative value, the road surface temperature after heating and cooling corresponding to the road surface to be monitored is determined, and the road surface temperature after heating and cooling includes the road surface temperature after heating or the road surface temperature after cooling.

[0130] In one embodiment, the measured meteorological data includes wind speed data and air temperature data; the multi-factor heat conversion formula includes a heat conduction energy conversion formula, a heat radiation energy conversion formula, and a convection heat transfer energy conversion formula;

[0131] The expression of heat conduction energy conversion formula is:

[0132] ;

[0133] in, is the energy value generated during heat conduction; For the moment; is the unit radiation road area; is the thermal conductivity of the material; For the road surface to be monitored at time Air temperature data below; For the road surface to be monitored at time Current road surface temperature under is the coordinate of the heat transfer direction;

[0134] The expression of thermal radiation energy conversion formula is:

[0135] ;

[0136] in, is the energy value generated during the thermal radiation process; For the moment; is the unit radiation road area; is solar radiation; For the Thermal characteristic wavelength absorptivity of each thermal characteristic wave; For the The proportion of thermal characteristic wave energy of each thermal characteristic wave;

[0137] The expression of the convective heat transfer energy conversion formula is:

[0138]

[0139] in, is the energy value generated during the convection heat transfer process; is the specific heat capacity of the material; For quality; For the road surface to be monitored at time The road surface temperature changes below; is the corrected initial road surface temperature; is the road surface temperature when convective heat transfer begins; For the road surface to be monitored at time Air temperature data below; is the standard deviation; is the mean; is the wind speed data; For the moment Wind speed data to time The duration when the wind speed value remains unchanged between the wind speed data.

[0140] In one embodiment, the parameter correction module 308 is specifically configured to:

[0141] For any parameter to be corrected in the training data set and the model parameters of the neural network model, the parameter is corrected based on the road surface temperature after heating and cooling and the measured road surface temperature value, and the other parameters except the parameter are fixed.

[0142] In one embodiment, the parameters to be calibrated include the specific heat capacity of the material in the training data set and the unit temperature rise representative value of the neural network model.

[0143] In one embodiment, a temperature monitoring module is further included for:

[0144] Obtain material property information and meteorological time series data corresponding to the road surface to be monitored;

[0145] Predict the wind speed and air temperature data of the road surface to be monitored at the monitoring time based on meteorological time series data;

[0146] The material property information, wind speed data and air temperature data of the road surface to be monitored at the monitoring time are input into the trained neural network model to monitor the temperature of the road surface to be monitored through the neural network model.

[0147] The device provided in the embodiment of the present invention has the same implementation principle and technical effects as those in the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference can be made to the corresponding content in the aforementioned method embodiment.

[0148] An embodiment of the present invention provides an electronic device. Specifically, the electronic device includes a processor and a storage device. The storage device stores a computer program, and when the computer program is executed by the processor, it executes the method described in any one of the above-mentioned embodiments.

[0149] Figure 4 This is a structural diagram of an electronic device provided in an embodiment of the present invention. The electronic device 100 includes: a processor 40, a memory 41, a bus 42 and a communication interface 43. The processor 40, the communication interface 43 and the memory 41 are connected via the bus 42; the processor 40 is used to execute an executable module stored in the memory 41, such as a computer program.

[0150] Memory 41 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between the system network element and at least one other network element is achieved through at least one communication interface 43 (which may be wired or wireless), and may utilize the Internet, a wide area network, a local area network, a metropolitan area network, or the like.

[0151] The bus 42 may be an ISA bus, a PCI bus, or an EISA bus. The bus may be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, Figure 4 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0152] Among them, the memory 41 is used to store programs, and the processor 40 executes the program after receiving the execution instruction. The method executed by the device for flow process definition disclosed in any embodiment of the above-mentioned embodiment of the present invention can be applied to the processor 40 or implemented by the processor 40.

[0153] Processor 40 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method may be completed by hardware integrated logic circuits or software instructions in processor 40. The above processor 40 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processing unit (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It may implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present invention may be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory 41 , and the processor 40 reads the information in the memory 41 and completes the steps of the above method in combination with its hardware.

[0154] The computer program product of the readable storage medium provided in the embodiment of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method described in the previous method embodiment. The specific implementation can be referred to the previous method embodiment and will not be repeated here.

[0155] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0156] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or replace some of the technical features therein with equivalents. Such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A road surface temperature monitoring method based on road meteorological analysis, characterized in that: include: Step 1: Obtain material property information, measured meteorological data, and measured road surface temperature values ​​corresponding to the road surface to be monitored to construct a training data set for a neural network model; wherein the custom layer of the neural network model is configured with a multi-factor heat conversion formula; Step 2, correcting the initial road surface temperature corresponding to the road surface to be monitored; Step 3: Determine the temperature change pattern of the road surface to be monitored and its corresponding road surface temperature after heating and cooling based on the corrected initial road surface temperature and the training data set using the multi-factor heat conversion formula configured in the neural network model, wherein the multi-factor heat conversion formula includes a heat conduction energy conversion formula, a heat radiation energy conversion formula, and a convection heat transfer energy conversion formula; Step 4: calibrating the training data set and the model parameters of the neural network model based on the road surface temperature after the heating and cooling process and the measured road surface temperature value, wherein the parameters to be calibrated include the material specific heat capacity in the training data set and the unit temperature rise representative value of the neural network model; Step 5: Based on the corrected training data set and the corrected neural network model, repeat steps 1 to 4 until the training of the neural network model is completed. The trained neural network model is used to monitor the temperature of the road surface to be monitored based on the material property information and meteorological time series data corresponding to the road surface to be monitored.

2. The road surface temperature monitoring method based on road meteorological analysis according to claim 1, characterized in that: Correcting the initial road surface temperature corresponding to the road surface to be monitored includes: If the current iteration process is the first iteration process, then correcting the initial road surface temperature corresponding to the road surface to be monitored based on the measured road surface temperature value; If the current iterative process is not the first iterative process, the initial road surface temperature corresponding to the road surface to be monitored is corrected based on the current road surface temperature of the road surface to be monitored at a specified time on the previous day output by the neural network model.

3. The road surface temperature monitoring method based on road meteorological analysis according to claim 1, characterized in that: Determining the temperature change pattern corresponding to the road surface to be monitored and its corresponding road surface temperature after heating and cooling based on the corrected initial road surface temperature and the training data set using the multi-factor heat conversion formula configured in the neural network model includes: determining an absolute absorbed energy value based on the corrected initial road surface temperature and the training data set using the multi-factor heat conversion formula configured in the neural network model; Determining a temperature change mode corresponding to the road surface to be monitored based on the positive or negative value of the absolute absorbed energy value, wherein the temperature change mode includes a heating mode or a cooling mode; The road surface temperature after the temperature increase or decrease corresponding to the road surface to be monitored is determined according to the temperature change pattern and the unit temperature increase representative value.

4. The road surface temperature monitoring method based on road meteorological analysis according to claim 3 is characterized in that: The measured meteorological data includes wind speed data and air temperature data; The heat conduction energy conversion formula is expressed as: ； in, is the energy value generated during heat conduction; For the moment; is the unit radiation road area; is the thermal conductivity of the material; For the road surface to be monitored at time Air temperature data below; For the road surface to be monitored at time Current road surface temperature under is the coordinate of the heat transfer direction; The thermal radiation energy conversion formula is expressed as: ； in, is the energy value generated during the thermal radiation process; For the moment; is the unit radiation road area; is solar radiation; For the Thermal characteristic wavelength absorption rate of each thermal characteristic wave; For the The proportion of thermal characteristic wave energy of each thermal characteristic wave; The expression of the convective heat transfer energy conversion formula is: in, is the energy value generated during the convection heat transfer process; is the specific heat capacity of the material; For quality; For the road surface to be monitored at time The road surface temperature changes below; is the corrected initial road surface temperature; is the road surface temperature when convective heat transfer begins; For the road surface to be monitored at time Air temperature data below; is the standard deviation; is the mean; is the wind speed data; For the moment Wind speed data to time The duration when the wind speed value remains unchanged between the wind speed data.

5. The road surface temperature monitoring method based on road meteorological analysis according to claim 1, characterized in that: Based on the road surface temperature after the heating and cooling process and the measured road surface temperature value, the training data set and the model parameters of the neural network model are corrected, including: For any parameter to be corrected in the training data set and the model parameters of the neural network model, the parameter is corrected based on the road surface temperature after heating and cooling and the measured road surface temperature value, and other parameters except the parameter are fixed.

6. The road surface temperature monitoring method based on road meteorological analysis according to claim 1, characterized in that: The method further comprises: Obtaining material property information and meteorological time series data corresponding to the road surface to be monitored; Predicting wind speed data and air temperature data of the road surface to be monitored at the monitoring time based on the meteorological time series data; The material property information, the wind speed data and the air temperature data of the road surface to be monitored at the monitoring time are input into the trained neural network model to monitor the temperature of the road surface to be monitored through the neural network model.

7. A road surface temperature monitoring device based on road meteorological analysis, characterized in that: include: A data set construction module is used to obtain material property information, measured meteorological data, and measured road surface temperature values ​​corresponding to the road surface to be monitored, so as to construct a training data set for the neural network model; wherein the custom layer of the neural network model is configured with a multi-factor heat conversion formula; An initial temperature correction module, used for correcting the initial road surface temperature corresponding to the road surface to be monitored; a post-change temperature prediction module, configured to determine a temperature change pattern corresponding to the monitored road surface and its corresponding road surface temperature after heating and cooling based on the corrected initial road surface temperature and the training data set using the multi-factor heat conversion formula configured in the neural network model, the multi-factor heat conversion formula including a heat conduction energy conversion formula, a heat radiation energy conversion formula, and a convection heat transfer energy conversion formula; a parameter correction module, configured to correct the training data set and model parameters of the neural network model based on the road surface temperature after the heating and cooling process and the measured road surface temperature value, wherein the parameters to be corrected include the material specific heat capacity in the training data set and the unit temperature rise representative value of the neural network model; A repeated execution module is used to repeat the data set construction module, the initial temperature correction module, the changed temperature prediction module, and the parameter correction module based on the corrected training data set and the corrected neural network model until the training of the neural network model is completed. The trained neural network model is used to monitor the temperature of the road surface to be monitored based on the material property information and meteorological time series data corresponding to the road surface to be monitored.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the method according to any one of claims 1 to 6.

Images