Asphalt paver virtual paving thickness real-time measurement method based on thickness measurement

Abstract

The invention relates to the field of mechanical engineering, and discloses an asphalt paver virtual paving thickness real-time measurement method based on thickness metering, and the method comprises the steps: collecting the surface height, local contact pressure and temperature data of a loose paving material in real time by synchronously deploying a laser ranging array, a contact pressure sensing array and a temperature sensor; and the embedded processing unit dynamically calculates the instantaneous rigidity of the material according to a pre-calibrated rigidity mapping table, corrects the geometric distance measurement value, and outputs the compensated virtual laying thickness. The system supports multi-level adaptive switching and sensor fault degradation output, and data continuity and reliability are ensured. Through multi-physics field coupling perception and a lightweight real-time compensation model, the thickness measurement stability and the process control precision are improved, and the equipment robustness and the engineering applicability are enhanced.

Description

Technical Field

[0001] This invention belongs to the field of mechanical engineering, specifically relating to a method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement. Background Technology

[0002] In modern road construction, asphalt concrete paving is a crucial process in road structure formation, and its construction quality directly determines the road's service life, driving comfort, and maintenance costs. The loose-lay thickness, a core technological parameter controlling the final thickness after compaction, affects not only the uniformity of the interlayer structure but also material usage control, cost accounting, and project acceptance standards. Therefore, achieving high-precision, real-time, and online measurement of loose-lay thickness has become an important direction for the evolution of intelligent road maintenance equipment technology. Early offline detection methods relying on manual ruler insertion or fixed-point sampling, due to their low efficiency, insufficient representativeness, and inability to provide closed-loop feedback, have been gradually replaced by automated sensing systems integrated into the paver itself, marking a fundamental shift in construction control from experience-driven to data-driven approaches.

[0003] Current mainstream technologies generally employ non-contact ranging sensors based on a single physical principle—such as ultrasonic sensors or laser displacement sensors—installed behind or to the side of the screed. By measuring the vertical distance from the sensor to the top surface of the loose paving material and combining this with a known reference surface height, the loose paving thickness is indirectly calculated. Under ideal conditions, this approach offers advantages such as rapid response, simple structure, and no mechanical wear, and has played a significant role in improving construction automation. Its technical logic is based on the assumption that the geometry of the loose paving surface can directly map thickness parameters, achieving process monitoring through high-frequency sampling, thus meeting the early basic needs for visualization and trend control. However, with the continuous improvement of high-grade highway construction standards, the increasing complexity of material formulations, and the stringent requirements for closed-loop accuracy in unmanned construction, the simplification inherent in this technical approach has encountered intractable technical problems.

[0004] The fundamental flaw of single-geometric ranging methods lies in their complete neglect of the dynamic disturbances to the measurement benchmark caused by the physical state of asphalt mixtures as heterogeneous viscoelastic-plastic media. Specifically, the temperature gradient, aggregate gradation variation, asphalt binder content fluctuation, and initial compaction differences during paving collectively determine the local stiffness and energy absorption characteristics of the material surface. For example, at high temperatures, asphalt viscosity decreases, resulting in a soft and porous layer. Ultrasonic waves attenuate and shift in phase due to impedance matching changes during penetration, while laser ranging suffers scattering errors due to surface undulations and thermal radiation interference. Conversely, at low temperatures or with high aggregate content, the surface of the mixture is rigid, and sensors may misinterpret local protrusions as an overall increase in thickness. Furthermore, different gradation designs (such as SMA and AC types) exhibit vastly different mechanical responses to the same sensor excitation due to differences in porosity and skeleton structure, causing the measured values ​​to systematically deviate from the true compaction potential. This measurement drift caused by the material's physical state is not random noise, but a structural error with strong working condition correlation. Its magnitude often exceeds the engineering tolerance, causing the control system to adjust the paving parameters based on distorted data, which in turn exacerbates the thickness dispersion and forms a negative feedback loop of measurement inaccuracy, improper control, and quality deterioration. Summary of the Invention

[0005] This invention aims to solve the technical problem in existing asphalt paving operations where the measurement of loose paving thickness is affected by operating parameters such as temperature, gradation, and density because a single geometric distance sensor cannot perceive the material's physical state. This leads to measurement reference drift, control inaccuracies, and increased dispersion of compacted thickness. Existing technologies rely on ultrasonic or laser displacement sensors to obtain the spatial coordinates of the loose paving surface. Their measurement logic is based on the simplified assumption that geometric shape is equivalent to compaction potential, ignoring the nonlinear differences in the response of asphalt mixtures as heterogeneous viscoelastic-plastic media to sensing excitation under different physical states. This invention constructs a multi-physics synchronous sensing architecture and a real-time stiffness compensation model to achieve a quantitative characterization of the instantaneous mechanical state of the material. Based on this, the geometric distance measurement results are dynamically corrected, thereby outputting stable, reliable, and directly applicable loose paving thickness data for closed-loop control of the compaction process.

[0006] As one embodiment of the present invention, the method is deployed in front of or to the side of the screed structure of an asphalt paver, and includes a non-contact ranging sensor array, a contact pressure sensor array, a temperature sensing module, an embedded data processing unit, and a communication interface module.

[0007] The non-contact distance sensor array consists of three laser displacement sensors, which are evenly spaced along the paving width. The sensors are installed at a height of 1.2m above the theoretical reference plane, with a measurement range of 0.3m to 1.5m, a repeatability of ±0.5mm, and a sampling frequency of 500Hz. They are used to obtain the absolute height data of the top surface of the loose paving material at multiple lateral positions.

[0008] The contact pressure sensor array consists of a 3×3 matrix structure composed of 9 piezoresistive thin-film pressure sensors. The effective sensing area of ​​the sensor is 20mm×20mm, the range is 0 to 50KPa, the nonlinearity error is less than 1%, and the hysteresis is less than 0.5%. It is installed at the end of an elastically extendable floating pressure foot. The initial compression of the pressure foot is set to 5mm, and the spring stiffness coefficient is 800N / m, ensuring that the sensor maintains a constant light pressure contact with the loose material surface without damaging the original structure of the material.

[0009] The temperature sensing module uses a platinum resistance temperature sensor, which is installed at the outlet of the paver's auger spreader. The temperature measurement range is 0 to 300℃, the accuracy is ±1℃, and the response time is less than 200ms. It is used to obtain the average temperature value of the mixture entering the paving area in real time.

[0010] Furthermore, the embedded data processing unit is an industrial-grade ARM Cortex-A53 architecture processor with a main frequency of 1.2 GHz, equipped with 512 megabytes of running memory and 4 gigabytes of storage space, running a real-time operating system, and having multi-threaded parallel processing capabilities.

[0011] This unit communicates with the paver's main controller via a CAN bus interface to acquire process parameters such as paving speed, auger speed, and vibration frequency. It also synchronously acquires raw data streams from a non-contact ranging sensor array, a contact pressure sensor array, and a temperature sensor module via an RS485 interface. The data acquisition period is set to 20ms, and all sensor data includes hardware-level timestamps, ensuring a time synchronization error of less than 1ms.

[0012] Furthermore, the embedded data processing unit executes a stiffness compensation algorithm model, which includes a data preprocessing submodule, a stiffness coefficient calculation submodule, a thickness correction submodule, and an output submodule. The data preprocessing submodule performs a moving average filter on the raw pressure data with a window length of 5 sampling points to eliminate instantaneous impact interference; it also performs outlier removal on the ranging data using... The criteria identify and replace outliers; zero-order hold interpolation is performed on the temperature data to ensure strict alignment with the pressure and distance data on the time axis.

[0013] The stiffness coefficient calculation submodule is based on the preprocessed pressure value. With temperature value Call the pre-stored stiffness mapping table This mapping table was established through offline calibration experiments, covering a temperature range of 120℃ to 180℃, a pressure range of 5KPa to 45KPa, and a stiffness coefficient. Defined as the ranging correction caused by a unit pressure change, in mm / kPa, the mapping table is stored in embedded memory as a two-dimensional array with a resolution of 5°C for temperature and 2kPa for pressure. It is calculated in real-time using bilinear interpolation. Combination corresponding value.

[0014] The thickness correction submodule will convert the original non-contact ranging value With stiffness coefficient Multiply to obtain the correction amount Final output value of virtual piling thickness ,in A height compensation constant of 1.2m is installed in the system. The output submodule groups and packages the corrected thickness data according to the paver's lateral position and sends it to the paver's main controller via the CAN bus at a frequency of 50 frames per second. At the same time, it is uploaded to the cloud monitoring platform via the Ethernet interface.

[0015] Furthermore, the stiffness mapping table The offline calibration method is as follows: In a laboratory environment, standard-graded AC-13 asphalt mixture specimens were prepared, with the initial compaction degree controlled at 85%. Tests were conducted at four temperature points: 120℃, 140℃, 160℃, and 180℃. At each temperature point, the specimens were placed in a constant temperature chamber for 30 minutes, then moved to a testing platform. A pressure sensor array and distance sensor, consistent with those used in engineering projects, were installed on top. Five constant pressure levels of 5KPa, 15KPa, 25KPa, 35KPa, and 45KPa were applied to the specimen surface using a hydraulic loading device, with each pressure level maintained for 10 seconds. The pressure sensor output values ​​were recorded simultaneously. Distance sensor output value and infrared thermometer readings .

[0016] In each group Under these conditions, the actual thickness of the specimen was manually measured using a high-precision vernier caliper. Repeat the measurement three times and take the average value. Stiffness coefficient. Through formula Calculations show that for each The experiment was repeated 5 times, and the results were taken. The arithmetic mean of the values ​​is used as the calibration value for the corresponding cell in the mapping table. After calibration, the mapping table is permanently written to the embedded memory and cannot be modified during field operation.

[0017] Furthermore, the floating pressure foot structure of the contact pressure sensor array includes a guide sleeve, a compression spring, a pressure sensor base, and a wear-resistant pressure head. The guide sleeve has an inner diameter of 25mm, an outer diameter of 30mm, a length of 60mm, and is made of hard anodized aluminum alloy.

[0018] The compression spring has a free length of 50mm, a working length of 45mm after compression, a wire diameter of 2mm, a mean diameter of 20mm, and 8 effective turns; the pressure sensor base is a cylindrical stainless steel component with a diameter of 24mm and a height of 10mm, with a central opening for leading out the sensor signal line.

[0019] The wear-resistant pressure head is a hemispherical structure made of polytetrafluoroethylene with a radius of curvature of 10mm and a surface roughness of Ra0.8μm.

[0020] The presser foot is fixed to the ironing plate support beam by a threaded connection. After installation, ensure that the bottom surfaces of all press heads are on the same horizontal plane with a height error of less than 0.1mm.

[0021] Furthermore, the laser emitter of the non-contact ranging sensor array has a wavelength of 650nm, a spot diameter of 3mm at a distance of 1m, an IP67 protection rating, and a die-cast aluminum alloy housing. It is mounted via a universal adjustable bracket with pitch adjustment capability of ±15 degrees and horizontal adjustment capability of ±10 degrees. After installation, the optical axis perpendicularity must be calibrated using a laser target to ensure that the angle between the measuring beam and the normal to the theoretical reference plane is less than 1 degree. The sensor is powered by 24V DC, consumes less than 3W, and outputs a 0-10V analog voltage, which is sampled by a 16-bit analog-to-digital converter and input to the embedded processing unit.

[0022] Furthermore, the embedded data processing unit executes a self-test program upon startup, including sensor communication link testing, memory integrity verification, and stiffness mapping table boundary value verification. If the zero-point drift of the pressure sensor exceeds ±0.5 kPa, or the reference value of the ranging sensor deviates from the factory calibration value by more than ±1 mm, an alarm is triggered and thickness output is paused until manual reset or sensor replacement. During system operation, the zero-point value of each sensor is automatically recorded every 10 minutes and compared with the initial value. If the drift exceeds a threshold, software compensation is initiated, and the compensation coefficient is written to non-volatile memory and applied in subsequent calculations.

[0023] Furthermore, the method supports adaptive compensation for multi-gradation materials. This is achieved by pre-storing independent mapping sub-tables for three typical gradations—AC-13, SMA-13, and OGFC-13—in the stiffness mapping table. A gradation selection switch is set on the paver's operating interface, and the switch signal is transmitted to the embedded processing unit via a digital input interface. The processing unit then calls the corresponding sub-table to perform stiffness calculations based on the currently selected gradation. During gradation switching, the system automatically clears the current data cache and reinitializes the filter state to ensure data continuity is not disturbed.

[0024] Furthermore, when sending thickness data, the output submodule adds a data quality flag. A value of 0 indicates that the current data is calculated based on complete sensor input; a value of 1 indicates that the pressure sensor is partially faulty and stiffness is estimated using only distance measurement data and temperature interpolation; and a value of 2 indicates that the temperature sensor is faulty and the previous valid temperature value is used for compensation. The paver's main controller determines whether to adopt this data for automatic leveling control based on the flag. If the flag is 1 or 2 for 5 consecutive frames, the control mode is downgraded, switching to manual intervention mode.

[0025] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0026] 1. By simultaneously collecting three types of physical quantities—geometric height, contact pressure, and material temperature—a multi-dimensional sensing capability for the instantaneous mechanical state of asphalt mixtures was constructed. This overcomes the limitation that a single distance sensor can only reflect surface morphology and solves the problem of measurement reference drift caused by fluctuations in material properties from a fundamental perspective.

[0027] 2. The established stiffness compensation model is based on offline calibration experimental data, with a clear mapping relationship. The calculation process does not require complex iteration or machine learning training, has low resource consumption, and can run in real time on an embedded platform. It meets the millisecond-level response requirements of paving operations and ensures that the control system obtains continuous, stable, and high-confidence thickness feedback signals.

[0028] 3. The pressure sensor array adopts a floating pressure foot structure, which ensures reliable contact with the surface of the loose material while avoiding excessive pressure that could damage the original structure of the material. The measured pressure value truly reflects the local support stiffness of the material. When combined with temperature parameters and input into the compensation model, the corrected thickness value is closer to the actual potential thickness of the material after compaction, thus improving the accuracy of process control.

[0029] 4. The system has a sensor fault diagnosis and data degradation output mechanism. When some sensors fail, it can still provide identifiable estimation data, ensuring the continuity of construction and avoiding the paralysis of the entire measurement system due to a single point of failure. This improves the robustness and availability of the equipment under harsh working conditions.

[0030] 5. Supports adaptive compensation for multi-grade materials. By pre-storing stiffness mapping sub-tables for different grades and using manual selection switches, the same hardware system can be adapted to multiple material formulations without replacing sensors or recalibrating. This reduces equipment maintenance costs and operational complexity, and enhances the system's engineering practicality and promotional value. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the overall technical solution architecture of the real-time measurement method for the loose paving thickness of asphalt pavers based on thickness measurement proposed in this invention.

[0032] Figure 2 This is a schematic diagram of the core principle framework of the multi-physics synchronous sensing and stiffness compensation model in this invention;

[0033] Figure 3 This is a logical flow diagram of the sensor data acquisition and preprocessing stage in this invention;

[0034] Figure 4 This is a flowchart illustrating the logical flow of the dynamic calculation of stiffness coefficient and thickness correction stage in this invention.

[0035] Figure 5 This is a schematic diagram of the multi-level interaction relationship and data flow between the paver control system and the cloud platform in this invention; Detailed Implementation

[0036] This invention provides a real-time measurement method for the loose paving thickness of asphalt pavers based on thickness measurement. Its core lies in constructing a multi-physics synchronous sensing architecture and a real-time stiffness compensation model to quantitatively characterize the instantaneous mechanical state of the asphalt mixture. Based on this, the geometric distance measurement results are dynamically corrected, resulting in stable, reliable, and directly usable loose paving thickness data for closed-loop control of the compaction process. This method is deployed in front of or to the side of the asphalt paver's screed structure and includes a non-contact distance sensor array, a contact pressure sensor array, a temperature sensor module, an embedded data processing unit, and a communication interface module. All hardware components and software algorithms work together to form a complete closed-loop measurement and control system with a cycle time of 20ms, ensuring that data acquisition, processing, and output meet the stringent real-time requirements of paving operations.

[0037] The non-contact ranging sensor array consists of three laser displacement sensors, evenly spaced along the paving width. The sensors are installed 1.2m above the theoretical reference plane, with a measurement range of 0.3m to 1.5m, a repeatability of ±0.5mm, and a sampling frequency of 500Hz. Each sensor operates independently, with a laser emitter wavelength of 650nm, a spot diameter of 3mm at a distance of 1m, an IP67 protection rating, and a housing made of die-cast aluminum alloy.

[0038] The sensor is mounted using a universal adjustable bracket, which allows for pitch adjustment of ±15 degrees and horizontal adjustment of ±10 degrees. After installation, the optical axis perpendicularity must be calibrated using a laser target to ensure that the angle between the measuring beam and the normal to the theoretical reference plane is less than 1 degree. The sensor is powered by 24V DC, with a power consumption of less than 3W. The signal output is a 0 to 10V analog voltage, which is sampled by a 16-bit analog-to-digital converter and then input to the embedded processing unit. This array is used to acquire the absolute height data of the top surface of the loose paving material at multiple lateral positions, providing the original geometric reference for subsequent thickness calculations.

[0039] The contact pressure sensing array consists of a 3×3 matrix structure composed of 9 piezoresistive thin-film pressure sensors. The effective sensing area of ​​the sensor is 20mm×20mm, the range is 0 to 50KPa, the nonlinearity error is less than 1%, and the hysteresis is less than 0.5%.

[0040] All sensors are mounted on the end of a resiliently extendable floating pressure foot. The initial compression of the pressure foot is set to 5mm, and the spring stiffness coefficient is 800N / m, ensuring that the sensor maintains constant, light pressure contact with the loose material surface without damaging the material's original structure. The floating pressure foot structure includes a guide sleeve, a compression spring, a pressure sensor base, and a wear-resistant pressure head.

[0041] The guide sleeve has an inner diameter of 25mm, an outer diameter of 30mm, and a length of 60mm. It is made of hard anodized aluminum alloy. The compression spring has a free length of 50mm, a working length of 45mm after compression, a wire diameter of 2mm, a mean diameter of 20mm, and 8 effective turns. The pressure sensor base is a cylindrical stainless steel component with a diameter of 24mm and a height of 10mm. It has a central opening for leading out the sensor signal line. The wear-resistant pressure head is a hemispherical structure made of polytetrafluoroethylene with a radius of curvature of 10mm and a surface roughness of Ra0.8μm.

[0042] The presser feet are fixed to the ironing plate support beam via threaded connections. After installation, ensure that the bottom surfaces of all presser heads are on the same horizontal plane with a height error of less than 0.1mm. This array is used to sense the local support force of the loose material surface on the sensor. This force value directly reflects the instantaneous stiffness characteristics of the material under the current temperature and gradation.

[0043] The temperature sensing module uses a platinum resistance temperature sensor, installed at the outlet of the paver's auger spreader. It has a temperature measurement range of 0 to 300℃, an accuracy of ±1℃, and a response time of less than 200ms. This module is used to acquire the average temperature of the asphalt mixture entering the paving area in real time. This temperature value is a key parameter affecting the viscoelastic-plastic behavior of asphalt mixtures and, together with pressure sensing data, serves as the core input to the stiffness compensation model.

[0044] The embedded data processing unit is an industrial-grade ARM Cortex-A53 processor with a clock speed of 1.2 GHz, equipped with 512 megabytes of RAM and 4 gigabytes of storage, running a real-time operating system and possessing multi-threaded parallel processing capabilities. This unit communicates with the paver's main controller via a CAN bus interface to acquire process parameters such as paving speed, auger speed, and vibration frequency; it also synchronously acquires raw data streams from the non-contact ranging sensor array, contact pressure sensor array, and temperature sensor module via an RS485 interface. The data acquisition cycle is set to 20 ms, and all sensor data is timestamped with hardware, ensuring a time synchronization error of less than 1 ms. This unit is the core of the entire system, responsible for executing the stiffness compensation algorithm model, which includes a data preprocessing submodule, a stiffness coefficient calculation submodule, a thickness correction submodule, and an output submodule.

[0045] Upon system startup, the embedded data processing unit first executes a self-test program. This program includes sensor communication link testing, memory integrity verification, and stiffness mapping table boundary value verification. If the zero-point drift of the pressure sensor exceeds ±0.5 kPa, or the reference value of the distance sensor deviates from the factory calibration value by more than ±1 mm, an alarm is triggered and thickness output is paused until manual reset or sensor replacement. During system operation, the zero-point values ​​of each sensor are automatically recorded every 10 minutes and compared with the initial values. If the drift exceeds a threshold, software compensation is initiated, and the compensation coefficient is written to non-volatile memory and applied in subsequent calculations. This mechanism ensures the stability of measurement accuracy during long-term operation.

[0046] After the system completes its self-test and enters normal operation, step S1 is executed: data acquisition and time synchronization. The embedded data processing unit polls the non-contact ranging sensor array, the contact pressure sensor array, and the temperature sensor module via an RS485 bus at a period of 20ms. Upon receiving the polling command, each sensor immediately sends its current sampled value along with a hardware timestamp.

[0047] After receiving all data packets, the data processing unit first verifies the continuity and consistency of timestamps to ensure that all sensor data are acquired within the same 20ms period. If a sensor data packet is lost or a timestamp is abnormal, the system records the event and initiates a data recovery mechanism. For pressure and temperature data, the previous valid value is used for zero-order hold; for ranging data, if a single point is lost, adjacent sensor data is used for linear interpolation. After completing data acquisition and preliminary verification, the raw data is sent to the data preprocessing submodule.

[0048] Step S2 of the execution method: Data preprocessing. The data preprocessing submodule processes the three types of raw data separately. For the 9 channels of raw pressure data output from the contact pressure sensor array, a moving average filter is performed with a window length of 5 sampling points. That is, the arithmetic mean of the pressure values ​​of the current point and the previous 4 historical points is calculated to eliminate high-frequency noise caused by paver vibration or instantaneous material impact. The filtered pressure data is marked as follows. to These correspond to the 9 positions in a 3×3 matrix.

[0049] For the three channels of raw altitude data output by the non-contact ranging sensor array, outlier removal is performed. The criterion is to calculate the average of 10 sampling points, including the current period and the previous 9 periods. and standard deviation If the current sample value satisfy If it is a wild value, it will be replaced with .

[0050] The processed height data is labeled as , , For a single temperature value output by the temperature sensing module Because of its low sampling frequency and slow changes, zero-order hold interpolation is performed, which copies and holds the value until the next temperature sampling cycle, ensuring that it has a corresponding temperature value in each 20ms processing cycle along with the pressure and distance data. Preprocessed data , It is sent to the stiffness coefficient calculation submodule.

[0051] Execution method step S3: Stiffness coefficient calculation. The stiffness coefficient calculation submodule is based on the preprocessed pressure values. With temperature value Call the pre-stored stiffness mapping table This mapping table was established through offline calibration experiments, covering a temperature range of 120℃ to 180℃ and a pressure range of 5KPa to 45KPa. Stiffness coefficient. Defined as the ranging correction caused by a unit pressure change, with units of mm / kPa. The mapping table is stored in embedded memory as a two-dimensional array with a resolution of 5°C for temperature and 2 kPa for pressure.

[0052] For each pressure sensor location Its corresponding stiffness coefficient Real-time calculation is performed using bilinear interpolation. The specific process is as follows: First, find the value corresponding to the current temperature in the mapping table. and pressure The four nearest grid points , , , ,in and for less than and greater than The nearest temperature grid point, and for less than and greater than The nearest pressure grid point. Then, respectively at and Linear interpolation of pressure on the temperature line yields... and Finally, in terms of temperature... and Perform linear interpolation to obtain the final result. .

[0053] The calculation process was performed in parallel at the locations of the nine pressure sensors, resulting in nine stiffness coefficients. to .

[0054]

[0055] in, , , , Grid points , , , The system supports adaptive compensation for multi-gradation materials by pre-storing independent mapping sub-tables for three typical gradations: AC-13, SMA-13, and OGFC-13, in the stiffness mapping table. A gradation selection switch is set on the paver's operating interface. The switch signal is transmitted to the embedded processing unit via a digital input interface. The processing unit then calls the corresponding sub-table to perform stiffness calculations based on the currently selected gradation. During gradation switching, the system automatically clears the current data cache and reinitializes the filter state to ensure uninterrupted data continuity.

[0056] Execution method step S4: Thickness correction and data fusion. The thickness correction submodule will convert the original non-contact ranging values... With stiffness coefficient Multiply to obtain the correction amount Because there are 3 distance sensors and 9 pressure sensors, spatial data matching is required. The system divides the paving width into 3 zones, each zone corresponding to 1 distance sensor and 3 pressure sensors below it. For the first... Each region ( =1,2,3), its correction amount The calculation is the arithmetic mean of the correction values ​​of the three pressure sensors in this area, i.e. .

[0057] Finally, the output value of the loose paving thickness in this area. ,in This represents the original value for non-contact ranging within this area. A height compensation constant of 1.2m is installed for the system. , , Outputs from the three pressure sensors in this area. , , These are the three corresponding stiffness coefficients for this region. This formula unifies geometric height, material stiffness, and installation reference, outputting a thickness value with clear physical meaning.

[0058] Corrected thickness data , , This represents the real-time loose paving thickness at three key transverse positions of the paver.

[0059] Step S5 of the execution method: Data output and quality identification. The output submodule groups and packages the corrected thickness data according to the paver's lateral position and sends it to the paver's main controller via the CAN bus at a frequency of 50 frames per second. Simultaneously, it uploads the data to the cloud monitoring platform via the Ethernet interface. A data quality identification bit is attached when sending the data.

[0060] The calculation logic for the flag bit is as follows: First, check if the data from all nine pressure sensors is valid. If all are valid and the temperature data is valid, the flag bit is set to 0. If one or more pressure sensor data are invalid (e.g., exceeding the measurement range or communication interruption), but the temperature data is valid, the system only uses the distance measurement data and temperature interpolation to estimate a global stiffness coefficient. and use this Correction is performed on all areas, at which point the flag bit is set to 1; if the temperature sensor data fails, the previous valid temperature value is used for compensation calculation, and the flag bit is set to 2.

[0061] The paver's main controller determines whether to adopt the data for automatic leveling control based on the flag bit. If the flag bit is 1 or 2 for 5 consecutive frames, the control mode is downgraded, switching to manual intervention mode. This mechanism ensures that the system can still provide identifiable estimation data even when some sensors fail, guaranteeing construction continuity.

[0062] Stiffness Mapping Table The offline calibration method is the core foundation of this invention. In a laboratory environment, standard-graded AC-13 asphalt mixture specimens were prepared, with the initial compaction degree controlled at 85%. Tests were conducted at four temperature points: 120℃, 140℃, 160℃, and 180℃. At each temperature point, the specimens were placed in a constant-temperature chamber for 30 minutes, and then moved to a testing platform, on which a pressure sensor array and distance sensor, consistent with those used in engineering applications, were installed.

[0063] A hydraulic loading device was used to apply five constant pressure levels (5 kPa, 15 kPa, 25 kPa, 35 kPa, and 45 kPa) to the surface of the specimen. Each pressure level was maintained for 10 seconds, and the output values ​​of the pressure sensor were recorded simultaneously. Distance sensor output value and infrared thermometer readings .

[0064] In each group Under these conditions, the actual thickness of the specimen was manually measured using a high-precision vernier caliper. Repeat the measurement three times and take the average value. Stiffness coefficient. It is calculated using the formula for each The experiment was repeated 5 times, and the results were taken. The arithmetic mean of the values ​​is used as the calibration value for the corresponding cell in the mapping table. After calibration, the mapping table is permanently written to the embedded memory and cannot be modified during field operation. This calibration process ensures the accuracy and reliability of the mapping table data, which is the fundamental guarantee for the effectiveness of the real-time compensation algorithm.

[0065] The system described in this invention tightly integrates hardware structure and software algorithms to form a highly robust and high-precision real-time measurement system. A non-contact ranging sensor array provides high-precision spatial geometric information, a contact pressure sensor array provides mechanical information reflecting material properties, and a temperature sensing module provides environmental status information. These three components are deeply fused through an embedded data processing unit. The stiffness compensation model, based on physical principles and experimental calibration, is computationally simple and efficient, requiring no complex iterations and fully meeting the real-time requirements of the embedded platform. The system's self-checking, self-compensation, fault diagnosis, and data degradation output mechanisms enable stable operation even in harsh construction site environments, improving the automation level of asphalt paving operations and the uniformity of the final pavement quality.

[0066] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the term "inclusion" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0067] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement, characterized in that, include: Non-contact distance sensor arrays, contact pressure sensor arrays, and temperature sensor modules are deployed in front of or to the side of the screed of the asphalt paver. The embedded data processing unit synchronously collects the top surface height data of the loose material output by the non-contact ranging sensor array, the local contact pressure data output by the contact pressure sensor array, and the mixed material temperature data output by the temperature sensor module at a 20ms cycle. Outlier removal is performed on the altitude data, moving average filtering is performed on the pressure data, and zero-order hold interpolation is performed on the temperature data. Based on the pre-stored stiffness mapping table And bilinear interpolation algorithm, based on the filtered pressure value Interpolated temperature value Calculate the stiffness coefficient corresponding to each pressure sensing location. ; Subtract the correction amount obtained by multiplying the height data by the corresponding area stiffness coefficient and pressure value, and add the system installation height compensation constant of 1.2m to obtain the output value of the loose paving thickness of each area in the paving width direction. The output values ​​of the paving thickness are grouped and packaged by region, and after adding data quality identifier bits, they are sent to the paver main controller via CAN bus.

2. The method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement according to claim 1, characterized in that, The non-contact ranging sensor array consists of three laser displacement sensors arranged at equal intervals along the paving width direction, installed at a height of 1.2m above the theoretical reference plane, with a measurement range of 0.3m to 1.5m; The contact pressure sensing array consists of a 3×3 matrix structure composed of 9 piezoresistive thin-film pressure sensors. The effective sensing area of ​​the sensors is 20mm×20mm, and the range is 0 to 50KPa. The temperature sensing module uses a platinum resistance temperature sensor, which is installed at the outlet of the spiral fabric distributor and has a temperature measurement range of 0 to 300°C.

3. The method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement according to claim 2, characterized in that, The contact pressure sensor array is installed at the end of the elastically extendable floating pressure foot, with an initial compression of 5mm and a spring stiffness coefficient of 800N / m. The floating pressure foot includes a guide sleeve, a compression spring, a pressure sensor base, and a wear-resistant pressure head; The guide sleeve has an inner diameter of 25mm, an outer diameter of 30mm, and a length of 60mm. The compression spring has a free length of 50mm, a working length of 45mm, a wire diameter of 2mm, a mean diameter of 20mm, and 8 effective coils. The wear-resistant pressure head is a polytetrafluoroethylene hemispherical structure with a radius of curvature of 10 mm and a surface roughness of Ra 0.8 μm.

4. The method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement according to claim 3, characterized in that, The embedded data processing unit is an ARM Cortex-A53 architecture processor with a main frequency of 1.2 GHz, equipped with 512 megabytes of running memory and 4 gigabytes of storage space, and runs a real-time operating system; Process parameters such as paving speed, auger speed, and vibration frequency are obtained through the CAN bus interface; raw data streams from sensors are synchronously acquired through the RS485 interface, and all data are timestamped with hardware.

5. The method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement according to claim 4, characterized in that, The stiffness mapping table It covers a temperature range of 120℃ to 180℃ and a pressure range of 5KPa to 45KPa, and is stored in a two-dimensional array with a temperature step of 5℃ and a pressure step of 2KPa. stiffness coefficient Defined as the distance correction caused by a unit pressure change, with units of mm / KPa; When performing bilinear interpolation, first locate the four grid points adjacent to the current temperature and pressure values, then perform linear interpolation on the temperature and pressure lines respectively to obtain the intermediate values, and finally perform interpolation on the temperature dimension to obtain the final stiffness coefficient.

6. The method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement according to claim 5, characterized in that, The stiffness mapping table pre-stores independent mapping sub-tables for three gradations: AC-13, SMA-13, and OGFC-13. The paver's operating interface has a gradation selection switch, and the switch signal is transmitted to the embedded processing unit via a digital input interface. The processing unit calls the corresponding sub-table to perform stiffness calculations based on the selected gradation; When switching gradations, the data buffer is automatically cleared and the filter state is reinitialized.

7. The method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement according to claim 6, characterized in that, The outlier removal uses 3 The criterion is to calculate the mean of 10 sampling points, including the current period and the previous 9 periods. with standard deviation If the current sample value satisfy Then replace with ; The moving average filter window has a length of 5 sampling points, and the arithmetic mean of the pressure values ​​of the current point and the previous 4 historical points is calculated. The zero-order hold interpolation replicates and holds the temperature value until the next sampling period.

8. The method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement according to claim 7, characterized in that, The paving width is divided into 3 zones, each zone corresponding to 1 distance sensor and 3 pressure sensors; No. Regional correction ; No. Area tiling thickness output value ; For the first Raw values ​​of non-contact ranging within the area A height compensation constant of 1.2m is installed for the system. , , Outputs from the three pressure sensors in this area. , , These are the three corresponding stiffness coefficients for this region.

9. The method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement according to claim 8, characterized in that, A value of 0 for the data quality flag indicates that the pressure and temperature data are complete and valid. A value of 1 indicates that the pressure sensor is partially faulty, and the average value of the stiffness coefficient of the effective pressure sensor is used for global correction. A value of 2 indicates that the temperature sensor has failed and the previous valid temperature value is used for compensation. If the flag bit is 1 or 2 for 5 consecutive frames, a control mode downgrade is triggered.

10. The method for real-time measurement of the loose paving thickness of asphalt pavers based on thickness measurement according to claim 9, characterized in that, The system executes a self-test program upon startup, including sensor communication link testing, memory integrity verification, and stiffness mapping table boundary value verification. If the zero point drift of the pressure sensor exceeds ±0.5KPa or the reference value of the distance sensor deviates from the factory calibration value by more than ±1mm, an alarm will be triggered and the thickness output will be paused. The sensor zero-point value is recorded every 10 minutes during operation. If the drift exceeds the threshold, software compensation is initiated and the compensation coefficient is written to non-volatile memory.

Images