Wideband adaptive tracking resonant compaction closed-loop control method for time-varying inherent frequency of asphalt mixture

Abstract

The present application relates to the field of intelligent compaction technology of road engineering, especially to a wideband adaptive tracking resonant compaction closed-loop control method for asphalt mixture time-varying natural frequency, which collects the vibration response signal of the steel wheel of the road roller and the temperature information and compaction state information of the asphalt mixture; based on the vibration response signal, the current time natural frequency of the asphalt mixture is identified online to obtain the natural frequency observation value as the filter tracking input; the natural frequency observation value, the natural frequency state quantity at the historical time, and the temperature information and compaction state information are jointly input into the filter prediction model to obtain the time-varying natural frequency estimation value; the target vibration frequency is determined according to the time-varying natural frequency estimation value, and the vibration amplitude is synchronously determined under the preset constraint condition. In the present application, the natural frequency matching degree of the whole construction cycle is greater than or equal to 95%, the compaction degree variation coefficient is reduced from 1.0% to 0.4%, the uniformity is improved by 60%, the aggregate crushing rate is reduced from 15% to 2.8% or less, and the pavement fatigue life is improved by 25%.

Description

Technical Field

[0001] This invention relates to a broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequency of asphalt mixtures, belonging to the field of intelligent compaction technology in road engineering. Background Technology

[0002] Resonant compaction is one of the most efficient compaction technologies currently available. By matching the vibration frequency of the roller to the natural frequency of the asphalt mixture, it achieves maximum compaction with minimal excitation force. However, existing methods have the following problems:

[0003] Traditional asphalt mixture compaction uses a fixed preset resonant frequency, meaning that under a single working condition, the entire compaction process is completed with a single vibration frequency. This cannot fully match the entire process of asphalt mixture from loose to fully compacted, nor can it adapt to the inherent time-varying characteristics of frequency caused by changes in asphalt temperature and compaction.

[0004] Currently, domestic brand road rollers have made some adjustments to the traditional compaction mode. They use FFT spectrum analysis to identify the natural frequency, reducing the identification delay to 50ms. However, when the temperature drops or there is strong wind, the natural frequency of the asphalt mixture changes by 2 to 3 Hz every 10 seconds. This is completely unable to keep up with the rate of frequency change, and under-compaction / over-compaction problems are likely to occur.

[0005] In addition, neither domestic nor international matching methods currently consider the problem of aggregate damage during resonant compaction. In fact, current road rollers in soft particle-graded asphalt pavements damage the aggregate in the mixture, which significantly reduces the fatigue life of the pavement. Therefore, it is urgent to improve the broadband adaptive tracking resonant compaction closed-loop control method of asphalt mixture time-varying natural frequency to solve the above-mentioned problems. Summary of the Invention

[0006] The purpose of this invention is to provide a broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequencies of asphalt mixtures, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequencies of asphalt mixtures includes the following steps:

[0009] The vibration response signal of the steel wheel of the road roller, as well as the temperature and compaction status information of the asphalt mixture, are collected.

[0010] Based on the vibration response signal, the natural frequency of the asphalt mixture at the current moment is identified online to obtain the natural frequency observation value as the input for filtering and tracking.

[0011] The natural frequency observation value, the natural frequency state quantity at historical time, and the temperature information and compaction state information are input into the filtering prediction model to obtain the time-varying natural frequency estimate.

[0012] The target vibration frequency is determined based on the time-varying natural frequency estimate, and the vibration amplitude is simultaneously determined under preset constraints.

[0013] The vibration parameters of the road roller are adjusted according to the target vibration frequency and vibration amplitude, and the filter prediction model and control parameters are updated cyclically based on the compaction feedback results to achieve closed-loop adaptive control of the asphalt mixture compaction process.

[0014] Furthermore, the online identification of the current natural frequency of the asphalt mixture based on the vibration response signal specifically includes:

[0015] Establish the state-space equations for the coupled dynamics of the steel wheel and the asphalt mixture:

[0016] ;

[0017] In the formula, The system state vector at time k contains the displacement, velocity, and acceleration modal components of the asphalt surface layer. The system state matrix, For the output matrix, , These are process noise and observation noise, respectively. The vibration response signal was collected at time k;

[0018] The recursive random subspace method is used to perform eigenvalue decomposition on the state matrix, and the current natural frequency of the asphalt mixture is identified in real time.

[0019] Furthermore, the real-time identification of the current inherent frequency of the asphalt mixture specifically includes:

[0020] According to the formula:

[0021] ;

[0022] The calculation has a fixed frequency. State matrix The i-th eigenvalue, The sampling interval is defined as ; and the first-order natural frequency with the highest actual contribution rate is extracted as the tracking target.

[0023] Furthermore, the step of inputting the observed natural frequency value, the natural frequency state quantity at historical times, and the temperature and compaction state information into the filtering prediction model to obtain the estimated time-varying natural frequency value specifically includes:

[0024] The extended Kalman filter model is used for broadband adaptive time-varying frequency tracking. Its state equation and observation equation are as follows:

[0025] ;

[0026] In the formula, Let k be the estimated time-varying natural frequency at time k. The temperature information of the asphalt mixture at time k is given. This represents the compaction state information at time k. These are the pre-calibrated temperature-frequency sensitivity coefficient and compaction-frequency sensitivity coefficient, respectively. These are natural frequency observations. These are process noise and observation noise, respectively.

[0027] The gain update formula for the extended Kalman filter is:

[0028] ;

[0029] in These are the process noise covariance and the observation noise covariance, respectively. The above model enables unbiased tracking of the natural frequency.

[0030] Further, the step of determining the target vibration frequency based on the estimated time-varying natural frequency and simultaneously determining the vibration amplitude under preset constraints specifically includes:

[0031] Establish a frequency-amplitude co-optimization function with aggregate crushing constraints:

[0032] ;

[0033] And it satisfies the following constraints:

[0034] ;

[0035] in, The target vibration frequency is determined at time k. For the synchronously determined vibration amplitude, This is an estimate of the time-varying natural frequency. This is the predicted value for aggregate crushing rate. The correlation coefficient is the gradation correlation coefficient. These are the weighting coefficients. This is the predicted compaction degree. The minimum compaction degree required by the design is determined by solving the optimization function, and the target vibration frequency and vibration amplitude are determined simultaneously under the premise of satisfying the constraints.

[0036] Furthermore, in the constraints, , , , , .

[0037] Furthermore, the step of cyclically updating the filtered prediction model and control parameters based on the compaction feedback results to achieve closed-loop adaptive control of the asphalt mixture compaction process specifically includes:

[0038] The entire process calculation is performed with a preset control cycle, and the steel wheel vibration response signal, asphalt mixture temperature information and compaction status information are collected synchronously in each cycle.

[0039] Based on the collected data, the recursive random subspace method is used to identify the natural frequency observation value at the current time.

[0040] The natural frequency observations are filtered using an extended Kalman filter to obtain smoothed time-varying natural frequency predictions.

[0041] Solve the constrained frequency-amplitude co-optimization function to obtain the optimal combination of target vibration frequency and vibration amplitude;

[0042] Control commands are issued to adjust the vibration parameters of the road roller, while compaction feedback data is collected. The temperature-frequency sensitivity coefficient, compaction degree-frequency sensitivity coefficient, and gradation correlation coefficient in the frequency-amplitude co-optimization function in the filtered prediction model are updated at preset time intervals to adapt to changes in working conditions.

[0043] Furthermore, the preset control cycle is 20ms, and the preset time interval is 5min.

[0044] Furthermore, the control method is implemented through a system comprising a sensing layer, including a triaxial accelerometer, an infrared temperature sensor, and a real-time compaction degree acquisition sensor mounted on the steel wheel of the road roller, for acquiring the vibration response signal of the steel wheel, the temperature information of the asphalt mixture, and the compaction status information;

[0045] The main control unit uses an industrial-grade chip to receive data collected by the sensing layer in real time at a preset period. It integrates a recursive random subspace identification module, an extended Kalman filter tracking module, and a constrained optimization solution module to perform online identification of natural frequencies, time-varying natural frequency filtering prediction, and collaborative optimization solution of target vibration frequency and vibration amplitude.

[0046] The control layer includes a vibration frequency control module and an amplitude frequency control module, which are used to receive the target vibration frequency and vibration amplitude output by the main control unit and drive the roller vibrator to adjust the vibration parameters.

[0047] The calibration coefficient iterative update module is used to update the temperature-frequency sensitivity coefficient, compaction degree-frequency sensitivity coefficient, and gradation correlation coefficient in the filtered prediction model at preset time intervals based on the compaction feedback data.

[0048] Further, it includes: a processor and a memory, the processor being connected to the memory, the memory being used to store a computer program, and the processor being used to execute the computer program stored in the memory, so that the electronic device executes the asphalt mixture time-varying natural frequency broadband adaptive tracking resonance compaction closed-loop control method.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] This invention achieves a natural frequency matching degree of ≥95% throughout the entire construction cycle, reduces the compaction degree variation coefficient from 1.0% to 0.4%, and improves uniformity by 60%; reduces aggregate breakage rate from 15% to within 2.8%, and increases pavement fatigue life by 25%; has a frequency tracking delay of ≤5ms, which is 90% lower than the traditional FFT scheme, and is fully adaptable to rapidly changing construction scenarios such as high-altitude and low-temperature environments; reduces the average number of compaction passes by 1.5 passes, improves construction efficiency by 20%, and reduces fuel consumption by 12%; requires no pre-embedded sensors on the pavement, only the addition of an acceleration sensor and main control unit to the existing road roller, with a modification cost of ≤15,000 yuan / unit, and supports compatibility with mainstream brand road rollers. Attached Figure Description

[0051] Figure 1 This is a system block diagram of the broadband adaptive tracking resonance compaction closed-loop control system for time-varying natural frequency of asphalt mixtures according to the present invention.

[0052] Figure 2 This is a flowchart of the time-varying natural frequency tracking algorithm of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Example 1

[0055] This embodiment takes the construction of the surface layer of the SMA-13 ​​expressway in a high-altitude and cold region as an example to illustrate in detail the specific implementation process of the broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequency of asphalt mixture of the present invention.

[0056] Hardware preparation and parameter calibration: A triaxial accelerometer with a sampling frequency of 2kHz, an infrared temperature sensor, and a real-time vibration compaction value acquisition system are installed on the steel drum of the road roller. The main control unit uses an STM32H743 industrial-grade chip, which supports real-time calculation with a 20ms cycle, a frequency adjustment resolution of 0.1Hz, and covers a wide frequency adjustment range of 20-60Hz.

[0057] Before formal construction, pre-calibration was conducted on a test section to obtain model parameters suitable for SMA-13 ​​asphalt mixture: temperature-frequency sensitivity coefficient η = 0.12 Hz / ℃; compaction degree-frequency sensitivity coefficient μ = 0.35 Hz / ℃; gradation correlation coefficient α = 1.2; and the minimum compaction degree required by design. ≥96%; aggregate crushing rate constraint The weighting coefficient is set to Prioritize frequency matching while limiting amplitude to prevent aggregate breakage.

[0058] Real-time data acquisition and initial state establishment

[0059] After construction began, the main control unit performed full-process calculations with a period of 20ms; at the initial moment (k=1), the following data were collected: asphalt mixture paving temperature T(1)=170℃, initial compaction degree K(1)=88%, and steel wheel vibration response signal. ;

[0060] Online identification of time-varying natural frequencies based on SSI

[0061] Treating the steel wheel and asphalt mixture as a viscoelastic half-space, we establish the coupled dynamic state-space equations: The recursive random subspace method is used to perform eigenvalue decomposition on the state matrix A. For the initial time, the first-order natural frequency observation is extracted: The initial natural frequency observation values ​​were obtained through calculation. The calculation takes about 1.8ms, which meets the real-time requirements.

[0062] Time-varying natural frequency wideband tracking based on EKF: The natural frequency is smoothly estimated using an extended Kalman filter model. The state equation and observation equation are as follows:

[0063] ;

[0064] Set initial state estimates Initial error covariance Process noise covariance Observation noise covariance .

[0065] Perform EKF gain update in each control cycle:

[0066] ;

[0067] Throughout the compaction process, the natural frequency changed continuously as the compaction degree increased and the temperature decreased. At 12 minutes of compaction (corresponding to k=36000 cycles), the collected data were as follows:

[0068] Asphalt temperature T=125℃, real-time compaction degree K=94%, SSI identification observation frequency The time-varying natural frequency estimate is obtained after EKF filtering. With a tracking latency of less than 5ms and a tracking error controlled within 0.2Hz, it is fully adapted to construction scenarios in cold and high-altitude regions where temperatures drop rapidly.

[0069] Frequency-amplitude co-optimization with aggregate crushing constraints:

[0070] Based on the time-varying natural frequency estimate from the EKF output Solve for the frequency-amplitude co-optimization function:

[0071] ;

[0072] The constraints are:

[0073] ;

[0074] In the initial stage of compaction (k=1, The optimal control parameters are obtained by solving the problem.

[0075] Target vibration frequency vibration amplitude Predicting aggregate crushing rate ;

[0076] Mid-compaction period (k=18000, The target vibration frequency is obtained by solving the problem. vibration amplitude Predicting aggregate crushing rate ;

[0077] Later stage of compaction (k=36000, The target vibration frequency is obtained by solving the problem. vibration amplitude Predicting aggregate crushing rate ;

[0078] Vibration parameter adjustment and closed-loop control

[0079] The main control unit will solve for the target vibration frequency and vibration amplitude The data is sent to the control layer, where the vibration frequency control module and amplitude frequency control module drive the roller vibrator to adjust the vibration parameters and achieve resonant compaction.

[0080] Meanwhile, the system continuously collects compaction feedback data. Every 5 minutes of cumulative operation, the main control unit iteratively updates the model parameters based on the compaction data of the most recent 5 minutes (including the correspondence between temperature changes and natural frequency changes, the correspondence between compaction degree increase and natural frequency increase, and the actual aggregate crushing rate test results).

[0081] Update the temperature-frequency sensitivity coefficient η, update the compaction-frequency sensitivity coefficient μ, and update the gradation correlation coefficient α;

[0082] By adaptively updating the parameters, the filter prediction model and optimization function continuously adapt to changes in different operating conditions, forming a complete closed-loop adaptive control.

[0083] Compaction effect test

[0084] After the compaction work was completed, the construction section was inspected, and the results are as follows:

[0085] Actual compaction degree: 97.2% (better than the design requirement of 96%), aggregate breakage rate: 2.7% (meets the constraint requirement of ≤3%), flatness standard deviation: 0.75mm, compaction degree variation coefficient: 0.4% (60% higher than the traditional compaction process of 1.0%), number of compaction passes: 1.5 fewer than the traditional process, improving construction efficiency by 20%.

[0086] Effect comparison and verification

[0087] To verify the technical advantages of this invention, a comparison group was set up in the same bidding section, using a traditional fixed-frequency compaction process. The comparison results are as follows:

[0088] index Traditional crafts This invention promote Frequency tracking delay 50ms ≤5ms Reduced by 90% Inherent frequency matching Approximately 70% ≥95% Increase by 25% Aggregate crushing rate 15% 2.8% Reduced by 81% coefficient of variation of compaction 1.0% 0.4% Reduce by 60% Construction efficiency benchmark Increase by 20% - Fuel loss benchmark Reduced by 12% -

[0089] Example 2

[0090] This embodiment was implemented in an ultra-thin wearing course construction scenario, using AC-10 asphalt mixture, a paving temperature of 160℃, and an initial compaction degree of 86%. Calibration was performed through a test section.

[0091] η = 0.15 Hz / ℃, μ = 0.40 Hz / %, gradation correlation coefficient α = 1.5;

[0092] Using the same control process as in Example 1, the inherent frequency tracking matching degree reached 96.2% throughout the process, the aggregate breakage rate was controlled within 2.5%, and the compaction uniformity was significantly improved, verifying the adaptability of the present invention under different mixture types and construction conditions.

[0093] Example 3

[0094] This embodiment was implemented in a high-altitude and low-temperature construction scenario, with an ambient temperature of -10℃ and a cooling rate of 0.8℃ / min after asphalt mixture paving. Using the method of this invention, the EKF tracking delay is ≤5ms, which can capture the rapid changes in the natural frequency in real time (change rate of about 0.5Hz / min), and the frequency matching degree throughout the process reaches 94.8%, effectively avoiding undervoltage and overvoltage problems. In contrast, the traditional FFT method (delay of 50ms) has a matching degree of only about 70% under the same conditions.

[0095] This application also provides a computer-readable storage medium storing a computer program that is executed by a processor to implement some or all of the steps of any of the device control methods described in the above method embodiments.

[0096] The inherent frequency matching degree throughout the entire construction cycle is ≥95%, the compaction degree variation coefficient is reduced from 1.0% to 0.4%, and the uniformity is improved by 60%; the aggregate crushing rate is reduced from 15% to less than 2.8%, and the road fatigue life is improved by 25%; the frequency tracking delay is ≤5ms, which is 90% lower than the traditional FFT scheme, and it is fully adaptable to rapidly changing construction scenarios such as high altitude and low temperature.

[0097] The average number of compaction passes is reduced by 1.5 passes, construction efficiency is increased by 20%, and fuel consumption is reduced by 12%. No sensors need to be pre-embedded in the road surface. Only an acceleration sensor and main control unit need to be added to the existing road roller. The modification cost is ≤15,000 yuan / unit. It is compatible with mainstream brand road rollers.

[0098] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0099] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequencies of asphalt mixtures, characterized in that: Includes the following steps: The vibration response signal of the steel wheel of the road roller, as well as the temperature and compaction status information of the asphalt mixture, are collected. Based on the vibration response signal, the natural frequency of the asphalt mixture at the current moment is identified online to obtain the natural frequency observation value as the input for filtering and tracking. The natural frequency observation value, the natural frequency state quantity at historical time, and the temperature information and compaction state information are input into the filtering prediction model to obtain the time-varying natural frequency estimate. The target vibration frequency is determined based on the time-varying natural frequency estimate, and the vibration amplitude is simultaneously determined under preset constraints. The vibration parameters of the road roller are adjusted according to the target vibration frequency and vibration amplitude, and the filter prediction model and control parameters are updated cyclically based on the compaction feedback results to achieve closed-loop adaptive control of the asphalt mixture compaction process.

2. The broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequency of asphalt mixture according to claim 1, characterized in that: The online identification of the current natural frequency of the asphalt mixture based on the vibration response signal specifically includes: Establish the state-space equations for the coupled dynamics of the steel wheel and the asphalt mixture: ； In the formula, The system state vector at time k contains the displacement, velocity, and acceleration modal components of the asphalt surface layer. The system state matrix, For the output matrix, , These are process noise and observation noise, respectively. The vibration response signal was acquired at time k. The recursive random subspace method is used to perform eigenvalue decomposition on the state matrix, and the current natural frequency of the asphalt mixture is identified in real time.

3. The broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequency of asphalt mixture according to claim 1, characterized in that: The real-time identification of the current inherent frequency of the asphalt mixture is specifically as follows: According to the formula: ； The calculation has a fixed frequency. State matrix The i-th eigenvalue, The sampling interval is defined as ; and the first-order natural frequency with the highest actual contribution rate is extracted as the tracking target.

4. The broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequency of asphalt mixture according to claim 1, characterized in that: The step of inputting the observed natural frequency value, historical natural frequency state variables, temperature information, and compaction state information into a filtering prediction model to obtain a time-varying natural frequency estimate specifically includes: The extended Kalman filter model is used for broadband adaptive time-varying frequency tracking. Its state equation and observation equation are as follows: ； In the formula, Let k be the estimated time-varying natural frequency at time k. The temperature information of the asphalt mixture at time k is given. This represents the compaction state information at time k. These are the pre-calibrated temperature-frequency sensitivity coefficient and compaction-frequency sensitivity coefficient, respectively. These are natural frequency observations. These are process noise and observation noise, respectively. The gain update formula for the extended Kalman filter is: ； in These are the process noise covariance and the observation noise covariance, respectively. The above model enables unbiased tracking of the natural frequency.

5. The broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequency of asphalt mixture according to claim 1, characterized in that: The step of determining the target vibration frequency based on the estimated time-varying natural frequency and simultaneously determining the vibration amplitude under preset constraints specifically includes: Establish a frequency-amplitude co-optimization function with aggregate crushing constraints: ； And it satisfies the following constraints: ； in, The target vibration frequency is determined at time k. For the synchronously determined vibration amplitude, This is an estimate of the time-varying natural frequency. This is the predicted value for aggregate crushing rate. The correlation coefficient is the gradation correlation coefficient. These are the weighting coefficients. This is the predicted compaction degree. The minimum compaction degree required by the design is determined by solving the optimization function, and the target vibration frequency and vibration amplitude are determined simultaneously under the premise of satisfying the constraints.

6. The broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequency of asphalt mixture according to claim 5, characterized in that: Among the constraints, , , , , .

7. The broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequency of asphalt mixture according to claim 1, characterized in that: The process of cyclically updating the filtered prediction model and control parameters based on the compaction feedback results to achieve closed-loop adaptive control of the asphalt mixture compaction process specifically includes: The entire process calculation is performed with a preset control cycle, and the steel wheel vibration response signal, asphalt mixture temperature information and compaction status information are collected synchronously in each cycle. Based on the collected data, the recursive random subspace method is used to identify the natural frequency observation value at the current time. The natural frequency observations are filtered using an extended Kalman filter to obtain smoothed time-varying natural frequency predictions. Solve the constrained frequency-amplitude co-optimization function to obtain the optimal combination of target vibration frequency and vibration amplitude; Control commands are issued to adjust the vibration parameters of the road roller, while compaction feedback data is collected. The temperature-frequency sensitivity coefficient, compaction degree-frequency sensitivity coefficient, and gradation correlation coefficient in the frequency-amplitude co-optimization function in the filtered prediction model are updated at preset time intervals to adapt to changes in working conditions.

8. The broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequency of asphalt mixture according to claim 1, characterized in that: The preset control cycle is 20ms, and the preset time interval is 5min.

9. The broadband adaptive tracking resonance compaction closed-loop control method for time-varying natural frequency of asphalt mixture according to any one of claims 1-8, characterized in that: The control method is implemented through the following system, which includes a sensing layer containing a triaxial acceleration sensor, an infrared temperature sensor, and a real-time compaction degree acquisition sensor installed on the steel wheel of the road roller, for collecting steel wheel vibration response signals, asphalt mixture temperature information, and compaction status information; The main control unit uses an industrial-grade chip to receive data collected by the sensing layer in real time at a preset period. It integrates a recursive random subspace identification module, an extended Kalman filter tracking module, and a constrained optimization solution module to perform online identification of natural frequencies, time-varying natural frequency filtering prediction, and collaborative optimization solution of target vibration frequency and vibration amplitude. The control layer includes a vibration frequency control module and an amplitude frequency control module, which are used to receive the target vibration frequency and vibration amplitude output by the main control unit and drive the roller vibrator to adjust the vibration parameters. The calibration coefficient iterative update module is used to update the temperature-frequency sensitivity coefficient, compaction degree-frequency sensitivity coefficient, and gradation correlation coefficient in the filtered prediction model at preset time intervals based on the compaction feedback data.

10. An electronic device, characterized in that, include: A processor and a memory, the processor being connected to the memory, the memory being used to store a computer program, the processor being used to execute the computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-8.

Images