Continuous paving and compaction detection and control method based on multi-dimensional vibration characteristics and closed loop control

Abstract

The present application relates to a kind of based on multi-dimensional vibration characteristics and closed loop control's continuous paving compaction detection and control method, it is related to compaction evaluation and control technical field.The present application is by the acceleration sensor installed on the steel wheel or frame of road roller real-time acquisition vibration acceleration signal, after pre-processing, calculate harmonic ratio, spectral entropy, impact response index and multi-dimensional compaction evaluation index such as vibration energy index, and establish the compaction degree mapping model between multi-dimensional index and semi-rigid base compaction degree, obtain continuous spatial compaction degree evaluation result.For multilayer continuous paving construction, construct space-time-level compaction quality matrix and interlayer coupling compensation model, according to the equivalent stiffness of lower layer, interlayer time interval and the interlayer state of semi-rigid base dynamically correct upper layer compaction degree target, and by PID and model predictive control adjust road roller driving speed, vibration frequency, amplitude and supplementary compaction number, realize compaction quality real-time detection, intelligent evaluation, construction cooperation and closed loop control.

Description

Technical Field

[0001] This invention relates to the field of road roller compaction assessment and control technology, and in particular to a continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control. Background Technology

[0002] Semi-rigid base courses, due to their high strength, good slab-like structure, and strong frost resistance, have become the main base course structure for high-grade highways and urban roads. The compaction quality of semi-rigid base courses directly affects the service life and performance of the pavement. Insufficient compaction leads to insufficient base course strength, uneven settlement, and premature damage, while over-compaction results in material breakage, poor interlayer bonding, and roller vibration. Therefore, real-time detection and precise control of the compaction quality of semi-rigid base courses are core technical issues in the field of road engineering.

[0003] Traditional compaction quality testing methods mainly rely on destructive testing methods such as sand cone method, ring cutter method, and core drilling method. These methods suffer from problems such as long testing cycles, sparse testing points, poor representativeness, and inability to provide real-time feedback, making it difficult to meet the requirements of modern road engineering for real-time control and full-process traceability of compaction quality. In recent years, intelligent compaction technology based on steel wheel vibration acceleration signals has gradually become a research hotspot in compaction quality testing. Its basic principle is to indirectly evaluate the stiffness state of the compacted material by analyzing the characteristic changes in the vibration signal of the road roller's steel wheel. For example, by analyzing the ratio of the first harmonic to the fundamental frequency of the steel wheel vibration acceleration signal, the nonlinear stiffness change of the compacted material can be reflected. These methods are simple in structure and easy to implement, but they suffer from insufficient description of the mechanical state of the compacted material using a single index, resulting in large evaluation errors; they are sensitive to noise and interference, exhibiting poor robustness; and they cannot distinguish the effects of changes in compaction degree and material type. In recent years, some studies have used machine learning methods (such as support vector machines, random forests, and neural networks) to establish mapping models between vibration characteristics and compaction degree. However, existing methods have the following shortcomings: poor model generalization ability, and the model performance drops significantly when the material type or construction conditions change; lack of uncertainty quantification, making it impossible to provide confidence information for engineering decisions; and poor model interpretability, making it difficult for engineers to understand the basis of the evaluation results.

[0004] In multi-layer continuous paving construction, factors such as the compaction quality of the lower layer, its age, and the interlayer condition of the semi-rigid base layer significantly affect the compaction control target of the upper layer. Current technologies primarily employ fixed compaction targets, neglecting the dynamic impact of interlayer coupling effects. This leads to unreasonable upper layer compaction control targets and difficulty in guaranteeing interlayer bonding quality. Furthermore, current technologies lack scheduling methods for coordinated operation of multiple rollers, resulting in operational conflicts and inefficiencies when multiple rollers are operating. Summary of the Invention

[0005] In order to solve the above-mentioned technical problems, or at least partially solve the above-mentioned technical problems, and in view of the above-mentioned shortcomings, the present invention provides a continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control.

[0006] In a first aspect, the present invention provides a method for continuous paving compaction detection and control based on multidimensional vibration characteristics and closed-loop control, comprising: During the compaction process of the road roller on each compaction unit, the vertical acceleration, longitudinal acceleration, and lateral acceleration of the road roller's steel wheel, as well as the road roller's position, speed, vibration frequency, and amplitude, are collected. The acquired signals are clock-synchronized and preprocessed. Within each time window or spatial grid, multidimensional compaction evaluation indices are calculated based on the preprocessed signal, including: harmonic ratio, normalized spectral entropy, impact response index, vibration energy index, and compaction uniformity index. The multidimensional compaction evaluation indexes, construction parameters, and interlayer state of the semi-rigid base layer of different compaction units are input into the compaction degree mapping model to obtain the current compaction degree estimate. The compaction degree mapping model adopts multiple regression, ridge regression, random forest, gradient boosting tree, XGBoost, or lightweight neural network model. When laying multiple layers in a continuous manner, the interlayer coupling compensation model is used to eliminate the interference of the lower layer state on the upper layer judgment, and the upper layer compaction target is corrected in order to obtain the corrected compaction target of each layer. Based on the estimated compaction degree and the compaction degree target for each level, control commands for the road roller are generated; the control commands are fed back to the road roller control system to achieve closed-loop control of compaction quality during multi-layer continuous paving construction.

[0007] Furthermore, the preprocessing includes outlier removal, mean removal, trend removal, bandpass filtering, wavelet denoising, and variational mode decomposition.

[0008] Furthermore, the harmonic ratio is calculated based on the fundamental frequency of the roller's steel wheel vibration and its higher harmonic energies. The process includes: performing a fast Fourier transform on the pre-processed acceleration signal of the roller's steel wheel to obtain the spectrum and further calculating the power spectrum; extracting the harmonic energy from the power spectrum, where the bandwidth of the m-th harmonic is Δ. f Then the energy of the mth harmonic is: ; Where q represents the total number of harmonics. Let x be the power spectral density amplitude at the k-th frequency point. The frequency corresponding to the k-th frequency point. For the fundamental frequency; Calculate the harmonic ratio using harmonic energy, using any of the following methods: Based on the fundamental wave, second and third harmonic energy The fundamental harmonic ratio is calculated as the final harmonic ratio, expressed as: ; in, To prevent extremely small positive numbers with a denominator of zero; To improve adaptability to different materials, rollers, and paving layers, the weighted harmonic ratio is calculated using the weighted harmonic energy of each order, and is used as the final harmonic ratio, expressed as: ; in, The weight of the m-th harmonic energy is determined based on the actual compaction response, or by adaptive harmonics to reduce the weight of higher harmonics. .

[0009] Furthermore, the calculation process of the normalized spectral entropy includes: performing a fast Fourier transform on the preprocessed acceleration signal of the road roller's steel wheel to obtain the spectrum, and further calculating the power spectrum; selecting a set of power spectrum points within the effective frequency band Ω containing the fundamental wave and the main harmonic from the power spectrum. Calculate the normalized spectral energy; calculate the spectral entropy based on the normalized spectral energy, and then normalize the spectral entropy.

[0010] Furthermore, the calculation process for the shock response index includes: Calculate the mean and standard deviation of the pre-processed roller steel wheel acceleration signal: Construct a set of shock events based on the mean and standard deviation: ; in, This is the threshold coefficient; The mean value of the acceleration signal of the steel wheel of the road roller after preprocessing is calculated. The standard deviation of the preprocessed roller steel wheel acceleration signal was calculated; among which, This indicates the pre-processed acceleration signal of the roller's steel wheel; shock events in the shock event set Construct an impact window centered on the target, and calculate the energy of a single impact as follows: , in, To impact the window; Sampling time interval; The maximum peak of the preprocessed signal within the impact window is taken as the single peak response. ; The energy-type impact response index is calculated based on the impact energy as the final impact response index: ; Alternatively, the peak-energy combined impact response index can be calculated based on the single peak response and impact energy: ; Where M represents the number of impact events. The reference impact energy and peak value are under standard compaction conditions. The weights are used to balance the contributions of the energy term and the peak term to the final index.

[0011] Furthermore, the calculation process for the compaction uniformity index includes: Calculate the mean value of the compaction degree estimates for the same compaction unit or adjacent compaction units; Using the mean and original value To calculate the coefficient of variation of the compaction estimate: , The compaction uniformity index is calculated using the coefficient of variation of the compaction degree estimate: .

[0012] Furthermore, an inter-layer coupling compensation model is constructed as follows: The equivalent stiffness index parameters of the lower layer are obtained based on the compaction mass matrix. The equivalent stiffness index of the lower layer is obtained by modeling the relationship between harmonic ratio, impact response index, vibration energy index and equivalent stiffness index using a multilayer perceptron. The compensation amount for the upper layer compaction target is calculated based on the lower layer's equivalent stiffness, interlayer time interval, and the interlayer state of the semi-rigid base layer: , in, The equivalent stiffness index of the lower layer l-1 of the l-th layer. Correction amount for the compaction target of the l-th layer To standardize and solidify targets, To achieve the target stiffness required by the design, The time interval between adjacent layers. For optimal moisture content, This represents the current water content of the layer. To compensate for the weighting coefficient, the compaction target of the upper layer is corrected using the compensation amount of the upper layer compaction target.

[0013] Furthermore, generating roller control commands based on the compaction degree estimate and the compaction degree targets at each level includes: calculating the compaction degree error based on the compaction degree estimate recorded in the compaction quality matrix and the corresponding compaction degree targets at each level; For the speed control variable, PID control is used for rapid adjustment. The speed change is calculated based on the compaction error using PID, and the next speed value is determined based on the speed change and the speed limit. , in, These are the proportional, integral, and differential coefficients; The compaction error is denoted by t and t-1.

[0014] Furthermore, generating roller control commands based on the estimated compaction degree and the compaction degree targets for each level includes: coupling control using a control objective function for multiple control variables such as frequency, amplitude, and number of compaction passes, and employing model predictive control.

[0015] Where J is the control objective function: ; Where H represents the prediction time domain, As weight, To predict the risk penalty term for segregation, vibration, or overvoltage at future prediction times in the time domain. To predict the error of future prediction times in the time domain, To predict the changes of multiple control variables at future prediction times in the time domain; constraints include: amplitude and frequency within a defined range, number of compaction cycles not exceeding the upper limit of the number of compaction cycles, and vibration oscillation not exceeding the upper limit of the vibration oscillation.

[0016] Furthermore, the quality grade is determined based on the estimated compaction degree and compaction uniformity index, and a supplementary compaction decision is made based on the grade determination result, including: Obtain the compaction quality grade for each compaction unit: ; Grade A indicates the estimated compaction degree. Meet the compaction target Furthermore, the compaction uniformity index meets the minimum threshold for compaction uniformity index. The compaction was satisfactory and no further compaction was required; however, the compaction degree of Grade B did not meet the target compaction degree. However, the compaction exceeds the level that requires an alarm. The required number of compaction passes is specified; Level C indicates that the compaction degree is less than the alarm compaction degree. Or the compaction uniformity index is lower than the minimum threshold for compaction uniformity index. However, if the uniformity is not up to standard, rework is required; During compaction, a compaction path is formed based on the priority of each compaction unit, and rollers are automatically allocated within the continuous paving construction window. The priority of each compaction unit is as follows: , in, This represents the maximum value of the sum of zero and the elements. Penalties for risks of segregation, vibration, or overvoltage. This refers to priority weights.

[0017] Secondly, the present invention provides a continuous compaction detection and control device based on multidimensional vibration characteristics and closed-loop control, comprising: at least one processing unit, wherein the processing unit is connected to a storage unit and an acquisition unit via a bus unit, the storage unit stores a computer program, and the processing unit implements the continuous compaction detection and control method based on multidimensional vibration characteristics and closed-loop control by running the computer program stored in the storage unit.

[0018] The technical solutions provided in the embodiments of the present invention have the following advantages compared with the prior art: This application preprocesses and extracts multidimensional compaction evaluation indicators by real-time acquisition of the acceleration signal of the steel wheel vibration of the road roller, comprehensively covering all compaction conditions and realizing continuous spatial detection of compaction quality. By establishing a compaction degree mapping model, a mapping relationship is established between the multidimensional compaction evaluation indicators and the compaction degree. By using the relationship between the multi-faceted evaluation indicators and the compaction degree to model, a quantitative evaluation of the compaction degree is realized, and the prediction results have good robustness.

[0019] This application uses interlayer coupling to set arbitrary upper layer compaction targets based on the equivalent stiffness of the lower layer. It considers the dynamic influence of interlayer coupling effects, solves the problem of the lower layer stiffness affecting the upper layer compaction setting, and ensures reasonable upper layer targets, thereby guaranteeing the upper layer compaction quality during continuous paving and improving the overall construction quality.

[0020] Based on the corrected compaction target and compaction degree estimate, the detection results are fed back to the roller control system in real time through closed-loop control, so as to realize the dynamic adjustment of speed, frequency, amplitude and number of passes. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart of a continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control provided in an embodiment of the present invention; Figure 2 A flowchart for calculating the harmonic ratio provided in an embodiment of the present invention; Figure 3This is a flowchart for calculating the normalized spectral entropy provided in an embodiment of the present invention; Figure 4 A flowchart for calculating the impact response index is provided for embodiments of the present invention; Figure 5 This is a flowchart for calculating the compaction uniformity index provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of a continuous paving compaction detection and control device based on multidimensional vibration characteristics and closed-loop control, provided in an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0025] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are 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. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0026] Example 1 like Figure 1 As shown, this invention provides a method for continuous paving compaction detection and control based on multidimensional vibration characteristics and closed-loop control, comprising: S100 collects the vertical acceleration, longitudinal acceleration, and lateral acceleration of the roller's steel wheel, as well as the roller's position, speed, vibration frequency, and amplitude during the compaction process of each compaction unit.

[0027] In the specific implementation process, each layer to be compacted is divided into multiple compaction units. The compaction unit is a grid of a set specification, and the signal length corresponding to the compaction unit is the number of sampling points corresponding to the time it takes for the road roller to pass through the compaction unit.

[0028] The system collects acceleration signals from the three axes of the road roller's steel wheel and simultaneously records parameters such as the road roller's position, speed, vibration frequency, and amplitude.

[0029] S200 performs clock synchronization alignment and preprocessing on the acquired signals; Preprocessing includes outlier removal, mean removal, trend removal, bandpass filtering, wavelet denoising, and variational mode decomposition.

[0030] Outlier removal employs three-sigma outlier removal. The DC component is removed by removing the mean, while the fluctuation component is retained. Due to sensor zero drift and low-frequency environmental interference, the original vibration signal contains a low-frequency trend term. This invention uses a least-squares polynomial fitting method to fit and remove the trend term, with the fitting order being 3 to 5. The detrended signal is the signal retaining only the fluctuations minus the polynomial fitting trend term.

[0031] The vibration signal contains periodic interference at known frequencies, such as engine vibration, hydraulic system vibration, and electrical vibration. This invention employs an adaptive notch filter to remove periodic interference at known frequencies and an adaptive bandpass filter to suppress out-of-band noise. The center frequency of the adaptive notch filter is dynamically adjusted according to the fundamental frequency of the road roller vibration. The passband range of the adaptive bandpass filter covers the fundamental frequency of the road roller vibration and its first five harmonics. After adaptive filtering, broadband noise may still exist in the signal. This invention uses db8 wavelets for 6-level decomposition and Bayes Shrink thresholding for soft thresholding. The Bayes Shrink threshold is adaptively determined based on the statistical characteristics of the wavelet coefficients at each level, effectively suppressing noise while preserving signal characteristics.

[0032] Vibration signals may contain multiple modal components, such as base response modes and noise modes. This invention employs variational mode decomposition (VMD) to decompose the signal into multiple intrinsic mode functions (EMFs), selecting the EMFs corresponding to the fundamental frequency and harmonics of the roller's steel wheel vibration as compaction-related vibration components. The number of modes K in the VMD is set to 4-6. VMD can accurately extract the fundamental and harmonic components directly related to the compaction state from complex vibration signals, improving the robustness of subsequent indicators. In other words, calculating multidimensional compaction evaluation indicators based on pure compaction-related vibration components effectively eliminates noise interference, making the evaluation results more stable and reliable.

[0033] S300. Within each time window or spatial grid, multidimensional compaction evaluation indices are calculated based on the preprocessed signal, including: harmonic ratio, normalized spectral entropy, impact response index, vibration energy index, and compaction uniformity index.

[0034] As the pavement is compacted, its stiffness increases, causing distortion in the waveform of the vibration wheel's acceleration signal, thus generating more high-order harmonics, which can be reflected by the harmonic ratio. For example... Figure 2 As shown, the harmonic ratio calculation process includes: The acceleration signal of the preprocessed road roller steel wheel was subjected to fast Fourier transform to obtain the spectrum and the power spectrum was further calculated. Harmonic energy is extracted from the power spectrum. Let the bandwidth of the m-th harmonic be Δf. Then the energy of the m-th harmonic is: ; Where q represents the total number of harmonics. Let x be the power spectral density amplitude at the k-th frequency point. The frequency corresponding to the k-th frequency point. With the fundamental frequency as the reference, the mth harmonic energy represents the total vibrational energy within a narrow band centered at m times the fundamental frequency.

[0035] The harmonic ratio is calculated using harmonic energy.

[0036] In one example, the fundamental harmonic ratio is calculated based on the fundamental, second, and third harmonic energies as the final harmonic ratio, which is expressed as: ; in, To prevent extremely small positive numbers with a denominator of zero.

[0037] To improve adaptability to different materials, rollers, and paving layers, the weighted harmonic ratio is calculated using the weighted harmonic energy of each order as the final harmonic ratio. The weighted harmonic ratio is expressed as: ; in, The weight of the m-th harmonic energy is determined based on the actual compaction response, or by adaptive harmonics to reduce the weight of higher harmonics. .

[0038] Normalized spectral entropy is used to measure the dispersion of vibration energy in the frequency domain, reflecting structural homogeneity and compaction stability. For example... Figure 3 As shown, the calculation process of normalized spectral entropy includes: The acceleration signal of the preprocessed road roller steel wheel is subjected to fast Fourier transform to obtain the spectrum and the power spectrum is further calculated. Select a set of power spectral points within the effective frequency band Ω that includes both the fundamental wave and the main harmonics from the power spectrum. The normalized spectral energy is: , For the first k , i Power spectral density amplitude at each frequency point; The spectral entropy is calculated based on the normalized spectral energy as follows: ; Using power spectrum point set The scale-normalized spectral entropy is: ; The closer the normalized spectral entropy is to 1, the more dispersed the spectral energy is, corresponding to strong dispersion of the semi-rigid base layer, uneven compaction, or unstable vibration response; the closer the normalized spectral entropy is to 0, the more concentrated the energy is, corresponding to a state where the soil is compacted and the system response tends to be stable.

[0039] The impact response index is used to describe the transient impact intensity, energy, and attenuation characteristics when a steel wheel comes into contact with a semi-rigid base layer. For example... Figure 4 As shown, the calculation process of the shock response index includes: Calculate the mean and standard deviation of the pre-processed roller steel wheel acceleration signal: ; in, This represents the nth discrete value of the preprocessed signal, where N is the signal quantity; Construct a set of shock events based on the mean and standard deviation: ; in, This is a threshold coefficient, typically taken as 2 to 3. The impact event needs to deviate from the mean. It is twice the standard deviation and not lower than the preceding and following signals.

[0040] shock events in the shock event set Construct an impact window centered on the target, and calculate the energy of a single impact as follows: , in, To impact the window; Sampling time interval; The maximum peak of the preprocessed signal within the impact window is taken as the single peak response: ; The energy-type impact response index is calculated based on the impact energy as the final impact response index: ; Alternatively, the peak-energy combined impact response index can be calculated based on the single peak response and impact energy as the final impact response index: ; Where M represents the number of impact events. The reference impact energy and peak value are under standard compaction conditions. The weights are used to balance the contributions of the energy term and the peak term to the final index.

[0041] A lower impact response index usually indicates that the semi-rigid base layer of the paving is loose or the steel wheel response is insufficient; a higher impact response index indicates greater local stiffness, aggregate concentration, segregation, or risk of vibration and jumping.

[0042] The vibration energy index reflects the intensity of the vibration response input from the road roller to the base layer per unit time or unit distance. The vibration energy index based on the sampling time interval is expressed as: , This represents the nth discrete value of the preprocessed signal.

[0043] If we consider the instantaneous travel speed of the road roller at the nth sampling time... The length L of the compaction unit is used, employing the vibration energy index per unit distance: .

[0044] The vibration energy index, based on the principle of energy conservation, aims to characterize the compaction state by analyzing the energy of vibration signals. The denser the semi-rigid base layer, the more vibration energy it typically absorbs and dissipates.

[0045] like Figure 5 As shown, the calculation process for the compaction uniformity index includes: Calculate the mean of the compaction degree estimates (DC) for the same compaction unit or adjacent compaction units: G represents the total amount of compaction calculated; if it is the same compaction unit, it is the compaction unit quantity.

[0046] The coefficient of variation was calculated using the mean and original values ​​of the compaction degree estimates: , The compaction uniformity index is calculated using the coefficient of variation of the compaction degree estimate: ; The closer the compaction uniformity index (UI) is to 1, the more uniform the spatial distribution of compaction quality.

[0047] The initial compaction estimate is the average compaction value under statistical uncompacted conditions.

[0048] S400 inputs the multidimensional compaction evaluation indexes, construction parameters, and interlayer state of the semi-rigid base layer of different compaction units into the compaction degree mapping model to obtain the compaction degree estimate of the current grid.

[0049] The interlayer conditions of semi-rigid base layers include kurtosis, skewness, material temperature, age, and moisture content. Construction parameters include driving speed, vibration frequency, amplitude, and number of compaction passes.

[0050] The compaction degree mapping model employs multiple regression, ridge regression, random forest, gradient boosting tree, XGBoost, or lightweight neural network models. Training data for the compaction degree mapping model comes from field test sections, including vibration signals and corresponding true values ​​of compaction degree using the sand cone method for different compaction passes (1-8 passes) and different material types (cement-stabilized crushed stone, lime-fly ash-stabilized crushed stone, and graded crushed stone). The training and test sets are divided in a 7:3 ratio. The training objective is to minimize the compaction degree error. Model validation metrics include the coefficient of determination (R²), root mean square error (RMSE), and mean relative error (MAPE).

[0051] In multi-layer continuous paving (S500), the vibration response of the upper layer is affected by the stiffness of the lower layer, resulting in deviations in the compaction degree mapped by the same vibration signal under different lower layer conditions. Therefore, it is necessary to formulate compaction degree targets for different layers. This application utilizes an inter-layer coupling compensation model to eliminate the interference of the lower layer condition on the judgment of the upper layer, so as to obtain the corrected compaction degree targets for each layer.

[0052] By dividing different construction surfaces into spatial grids formed by compaction units, a spatial-temporal-layer compaction quality matrix is ​​established using the output compaction estimates, multidimensional compaction evaluation indicators, and number of compaction passes. This matrix quantifies the impact of lower-layer compaction quality, equivalent stiffness, interlayer time interval, and interlayer state of semi-rigid base courses on the upper-layer compaction control objectives.

[0053] The compaction quality matrix includes compaction estimates at different locations, levels, and times, multidimensional compaction evaluation indices, and the number of compaction passes; the compaction quality matrix elements at location xy, time t, and level l are: ; in, This represents the compaction degree estimates at location xy, time t, and layer l using the compaction degree mapping model. Let x represent the number of compaction operations at position xy, time t, and level l; where... l Indicates hierarchy, l =1 indicates the bottom layer. l =2 indicates the second continuous layer. l =3 indicates the third consecutive layer.

[0054] The inter-layer coupling compensation model is constructed as follows: The parameters for calculating the equivalent stiffness index of the lower layer are obtained based on the compaction mass matrix. The equivalent stiffness index of the lower layer is obtained by modeling the relationship between the harmonic ratio, impact response index, vibration energy index and equivalent stiffness index using a multilayer perceptron.

[0055] The compensation amount for the upper layer compaction target is calculated based on the lower layer's equivalent stiffness, interlayer time interval, and the interlayer state of the semi-rigid base layer: , in, The equivalent stiffness index of the lower layer l-1 of the l-th layer. Correction amount for the compaction target of the l-th layer To standardize and solidify targets, To achieve the target stiffness required by the design, The time interval between adjacent layers. For optimal moisture content, This represents the current water content of the layer. The compensation weighting coefficient is used to compensate for the lower layer's undercompression. The compensation mechanism of the interlayer coupling compensation model can avoid misjudging the compaction target of the upper layer due to the lower layer's undercompression, and can also reduce the risk of upper layer vibration and segregation caused by the lower layer being too hard.

[0056] The compaction target of the upper layer is corrected by using the compensation amount of the upper layer compaction target.

[0057] S600: Generate roller control commands based on the compaction degree estimate and the compaction degree target for each layer; feed the control commands back to the roller control system to achieve closed-loop control of compaction quality during multi-layer continuous paving construction.

[0058] The control variables for closed-loop control of a road roller are defined as travel speed, vibration frequency, amplitude, number of compaction passes or number of additional compaction passes.

[0059] The compaction error is calculated based on the compaction degree estimates recorded in the compaction quality matrix and the corresponding compaction targets for each level.

[0060] For the speed control variable, PID control is used for rapid adjustment. The speed change is calculated based on the compaction error using PID, and the next speed value is determined based on the speed change and the speed limit. , in, These are the proportional, integral, and differential coefficients; The compaction error is denoted by t and t-1.

[0061] Within the prediction time domain, multiple control variables such as frequency, amplitude, and number of compaction cycles are coupled and controlled using a control objective function, employing model predictive control:

[0062] Where J is the control objective function: ; Where H represents the prediction time domain, As weight, To predict the risk penalty term for segregation, vibration, or overvoltage at future prediction times in the time domain. This is the error in predicting future prediction times in the time domain. To predict the changes of multiple control variables at future prediction times in the time domain; constraints include: amplitude and frequency within a defined range, number of compaction cycles not exceeding the upper limit of the number of compaction cycles, and vibration oscillation not exceeding the upper limit of the vibration oscillation.

[0063] In the specific implementation process, the quality grade is determined according to the estimated compaction degree and the compaction uniformity index, and the decision on additional compaction is made based on the grade determination result, including: Obtain the compaction quality grade for each compaction unit: ; Grade A indicates that the compaction degree meets the standard and has good uniformity, passing the acceptance test and requiring no additional compaction; Grade B indicates that the compaction degree does not meet the standard but exceeds the compaction degree that requires an alarm, requiring additional compaction a set number of times; Grade C indicates that the compaction degree is less than the alarm compaction degree. Or the compaction uniformity index is lower than the minimum threshold for compaction uniformity index. If the uniformity is not up to standard, rework is required.

[0064] During the compaction process, a compaction path is formed based on the priority of each compaction unit.

[0065] The priority of each compaction unit is as follows:

[0066] in, This represents the maximum value of the sum of zero and the elements. Penalties for risks of segregation, vibration, or overvoltage. This refers to priority weights.

[0067] The compaction path is generated based on priority, and the roller is automatically assigned within the continuous paving construction window.

[0068] When the impact response index exceeds the threshold, vibration is detected, and the vibration frequency or amplitude is immediately reduced. If necessary, vibration is stopped, and static pressure mode is adopted. When the compaction degree of three consecutive compaction units exceeds the target value by more than 2%, the system determines that overpressure has occurred and automatically skips this area to avoid material damage due to overpressure.

[0069] Example 2 See Figure 6As shown, this embodiment of the invention provides a continuous-laying compaction detection and control device based on multi-dimensional vibration characteristics and closed-loop control, comprising: at least one processing unit, wherein the processing unit is connected to a storage unit and a data acquisition unit via a bus unit, the storage unit serving as a computer-readable storage medium, and may be used to store software programs, computer-executable programs, and modules, such as the software program, computer-executable program, and module corresponding to the continuous-laying compaction detection and control method based on multi-dimensional vibration characteristics and closed-loop control in this embodiment of the invention. The processing unit implements the aforementioned continuous-laying compaction detection and control method based on multi-dimensional vibration characteristics and closed-loop control by running the software program, computer-executable program, and module stored in the storage unit.

[0070] Of course, the computer program stored in the storage unit of the continuous paving compaction detection and control system based on multidimensional vibration characteristics and closed-loop control provided in the embodiments of the present invention is not limited to the operation of the method described above, and can also execute related operations in the continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control provided in any embodiment of the present invention.

[0071] Example 3 This invention provides a computer-readable storage medium storing a computer program. When the computer program is executed, it implements the continuous compaction detection and control method based on multidimensional vibration characteristics and closed-loop control.

[0072] The computer-readable storage medium provided in the embodiments of the present invention stores a computer program that is not limited to the operation of the method described above, but can also execute related operations in the continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control provided in any embodiment of the present invention.

[0073] In the embodiments provided by this invention, it should be understood that the disclosed structures and methods can be implemented in other ways. For example, the structural embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, structures, or units, and may be electrical, mechanical, or other forms.

[0074] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0075] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0076] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for continuous paving compaction detection and control based on multidimensional vibration characteristics and closed-loop control, characterized in that, include: During the compaction process of the road roller on each compaction unit, the vertical acceleration, longitudinal acceleration, and lateral acceleration of the road roller's steel wheel, as well as the road roller's position, speed, vibration frequency, and amplitude, are collected. The acquired signals are clock-synchronized and preprocessed. Within each time window or spatial grid, multidimensional compaction evaluation indices are calculated based on the preprocessed signal, including: harmonic ratio, normalized spectral entropy, impact response index, vibration energy index, and compaction uniformity index. The multidimensional compaction evaluation indexes, construction parameters, and interlayer state of the semi-rigid base layer of different compaction units are input into the compaction degree mapping model to obtain the current compaction degree estimate. The compaction degree mapping model adopts multiple regression, ridge regression, random forest, gradient boosting tree, XGBoost, or lightweight neural network model. When laying multiple layers in a continuous manner, the interlayer coupling compensation model is used to eliminate the interference of the lower layer state on the upper layer judgment, and the upper layer compaction target is corrected in order to obtain the corrected compaction target of each layer. Based on the estimated compaction degree and the compaction degree target for each level, control commands for the road roller are generated; the control commands are fed back to the road roller control system to achieve closed-loop control of compaction quality during multi-layer continuous paving construction.

2. The continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control according to claim 1, characterized in that, The preprocessing includes outlier removal, mean removal, trend removal, bandpass filtering, wavelet denoising, and variational mode decomposition.

3. The continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control according to claim 1, characterized in that, The harmonic ratio is calculated based on the fundamental frequency of the roller's steel wheel vibration and its higher harmonic energies. The process includes: performing a fast Fourier transform on the pre-processed acceleration signal of the roller's steel wheel to obtain the spectrum and further calculating the power spectrum; extracting the harmonic energy from the power spectrum, where the bandwidth of the m-th harmonic is Δ. f Then the energy of the mth harmonic is: ； Where q represents the total number of harmonics. Let x be the power spectral density amplitude at the k-th frequency point. The frequency corresponding to the k-th frequency point. For the fundamental frequency; Calculate the harmonic ratio using harmonic energy, using any of the following methods: Based on the fundamental wave, second and third harmonic energy The fundamental harmonic ratio is calculated as the final harmonic ratio, expressed as: ； in, To prevent extremely small positive numbers with a denominator of zero; To improve adaptability to different materials, rollers, and paving layers, the weighted harmonic ratio is calculated using the weighted harmonic energy of each order, and is used as the final harmonic ratio, expressed as: ； in, The weight of the m-th harmonic energy is determined based on the actual compaction response, or by adaptive harmonics to reduce the weight of higher harmonics. .

4. The continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control according to claim 1, characterized in that, The calculation process of normalized spectral entropy includes: performing a fast Fourier transform on the preprocessed acceleration signal of the road roller's steel wheel to obtain the spectrum, and further calculating the power spectrum. From the power spectrum, a set of power spectrum points within the effective frequency band Ω containing the fundamental wave and the main harmonics is selected. Calculate the normalized spectral energy; calculate the spectral entropy based on the normalized spectral energy, and then normalize the spectral entropy.

5. The continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control according to claim 1, characterized in that, The calculation process of the shock response index includes: Calculate the mean value of the pre-processed roller steel wheel acceleration signal and standard deviation : Construct a set of shock events based on the mean and standard deviation: ； in, This is the threshold coefficient; This indicates the pre-processed acceleration signal of the roller's steel wheel; shock events in the shock event set Construct an impact window centered on the target, and calculate the energy of a single impact as follows: ， in, To impact the window; Sampling time interval; The maximum peak of the preprocessed signal within the impact window is taken as the single peak response. ; The energy-type impact response index is calculated based on the impact energy as the final impact response index: ； Alternatively, the peak-energy combined impact response index can be calculated based on the single peak response and impact energy: ； Where M represents the number of impact events. The reference impact energy and peak value are under standard compaction conditions. The weights are used to balance the contributions of the energy term and the peak term to the final index.

6. The continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control according to claim 1, characterized in that, The calculation process for the compaction uniformity index includes: Calculate the mean value of the compaction degree estimates for the same compaction unit or adjacent compaction units; Using the mean of the compaction degree estimates and original value To calculate the coefficient of variation of the compaction estimate: ， The compaction uniformity index is calculated using the coefficient of variation of the compaction degree estimate: .

7. The continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control according to claim 1, characterized in that, The inter-layer coupling compensation model is constructed as follows: The equivalent stiffness index parameters of the lower layer are obtained based on the compaction mass matrix. The equivalent stiffness index of the lower layer is obtained by modeling the relationship between harmonic ratio, impact response index, vibration energy index and equivalent stiffness index using a multilayer perceptron. The compensation amount for the upper layer compaction target is calculated based on the lower layer's equivalent stiffness, interlayer time interval, and the interlayer state of the semi-rigid base layer: ， in, The equivalent stiffness index of the lower layer l-1 of the l-th layer. Correction amount for the compaction target of the lth layer To standardize and solidify targets, To achieve the target stiffness required by the design, The time interval between adjacent layers. For optimal moisture content, This represents the current water content of the layer. To compensate for the weighting coefficient, the compaction target of the upper layer is corrected using the compensation amount of the upper layer compaction target.

8. The continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control according to claim 1, characterized in that, Generating roller control commands based on the compaction degree estimate and compaction degree targets at each level includes: calculating the compaction degree error based on the compaction degree estimate recorded in the compaction quality matrix and the corresponding compaction degree targets at each level; For the speed control variable, PID control is used for rapid adjustment. The speed change is calculated based on the compaction error using PID, and the next speed value is determined based on the speed change and the speed limit. ， in, These are the proportional, integral, and differential coefficients; The compaction error is denoted by t and t-1.

9. The continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control according to claim 1, characterized in that, Based on the estimated compaction degree and the compaction degree targets for each level, control commands for the road roller are generated, including: for multiple control variables such as frequency, amplitude, and number of compactions, coupled control is performed using a control objective function, employing model predictive control. Where J is the control objective function: ； Where H represents the prediction time domain, As weight, To predict the risk penalty term for segregation, vibration, or overvoltage at future prediction times in the time domain, To predict the error of future prediction times in the time domain, To predict the changes of multiple control variables at future prediction times in the time domain; constraints include: amplitude and frequency within a defined range, number of compaction cycles not exceeding the upper limit of the number of compaction cycles, and vibration oscillation not exceeding the upper limit of the vibration oscillation.

10. The continuous paving compaction detection and control method based on multidimensional vibration characteristics and closed-loop control according to claim 1, characterized in that, The quality grade is determined based on the estimated compaction degree and compaction uniformity index. Based on the grade determination, a decision on additional compaction is made, including: Obtain the compaction quality grade for each compaction unit: ； Grade A indicates the estimated compaction degree. Meet the compaction target Furthermore, the compaction uniformity index meets the minimum threshold for compaction uniformity index. The compaction was satisfactory and no further compaction was required; however, the compaction degree of Grade B did not meet the target compaction degree. However, the compaction exceeds the level that requires an alarm. The required number of compaction passes is specified; Level C indicates that the compaction degree is less than the alarm compaction degree. Or the compaction uniformity index is lower than the minimum threshold for compaction uniformity index. However, if the uniformity is not up to standard, rework is required; During compaction, a compaction path is formed based on the priority of each compaction unit, and rollers are automatically allocated within the continuous paving construction window. The priority of each compaction unit is as follows: ， in, This represents the maximum value of the sum of zero and the elements. Penalties for risks of segregation, vibration, or overvoltage. This refers to priority weights.

Images