Method and system for continuously detecting and evaluating subgrade compaction quality by vibration wheel response amplitude

Abstract

The present application relates to a kind of based on the continuous detection evaluation method and system of subgrade compaction quality of vibration wheel response frequency range, belong to subgrade construction field.The present application is by real-time detection vibration wheel vertical acceleration signal, time domain analysis is carried out to acceleration, and the acceleration energy spectrum is obtained, according to the size of frequency corresponding energy amplitude to determine vibration wheel response frequency range, on the basis, propose new index of subgrade filling compaction quality continuous detection-vibration response amplitude (Response value, RV), establish the relationship of RV and subgrade filling conventional compaction quality index.During subgrade rolling process, the three-dimensional space position of road roller is detected in real time by satellite positioning equipment, and the RV of subgrade full rolling surface is obtained, and the real-time, continuous, full coverage detection of subgrade compaction quality is realized.The continuous detection index of compaction quality proposed in the present application belongs to, and is suitable for the continuous detection and evaluation of the compaction quality of various subgrade fillings.

Description

Technical Field

[0001] This invention relates to an evaluation method and system, and more particularly to a method and system for continuous detection and evaluation of roadbed compaction quality based on the vibration wheel response amplitude, belonging to the field of roadbed construction. Background Technology

[0002] The roadbed is a crucial structural component of transportation engineering projects such as highways and railways, bearing its own weight, the weight of the pavement, and the loads from the superstructure. Its filling quality is critical to the long-term safe operation of the roadbed structure. The roadbed is primarily formed by vibratory roller compaction. Inadequate or uneven compaction can lead to a loose roadbed, causing surface cracking, voids, ground subsidence, and other problems such as frost heave and mud pumping under long-term static and dynamic loads. This results in poor service conditions and compromises traffic safety. Therefore, it is essential to strengthen the control of the roadbed compaction process and to inspect and control the compaction quality. Currently, roadbed compaction quality testing uses manual single-point sampling methods, which are time-dependent and highly susceptible to human influence. This method cannot achieve process control and comprehensive inspection of roadbed compaction quality, potentially creating safety hazards. Furthermore, the testing is time-consuming and inefficient, hindering construction progress control.

[0003] In the 1970s, European scholars were the first to conduct research on continuous vibration compaction testing technology. Thurner, from the Swedish Road Administration, installed acceleration sensors on vibratory rollers and first discovered that the ratio of the second harmonic amplitude to the fundamental amplitude in the acceleration frequency domain was related to the compaction state of the fill material. He proposed using the ratio of the second harmonic amplitude to the fundamental amplitude to evaluate the stiffness of the fill material and defined this value as the compaction value (CMV). In practice, it was found that not only the second harmonic exists in the frequency domain of vibratory roller acceleration, but also higher harmonics and fractional harmonics. Based on this, different scholars proposed various acceleration frequency domain indices, such as the Total Harmonic Distortion (THD) considering all integer harmonics, the Compaction Control Value (CCV) comprehensively considering the effects of integer and fractional harmonics, and the Resonance Value (RMV) considering the half-harmonic.

[0004] Based on this, some scholars have proposed a continuous detection method for roadbed compaction quality based on acceleration frequency domain analysis. For example, Chinese patent application number CN202011117230, publication number CN 112127342 A discloses a method for monitoring roadbed compaction quality based on spectrum and amplitude. The method mainly includes: installing vibration sensors on the roller of a road roller; the vibration sensors collect vibration signals of the road roller in real time and send them to the acquisition terminal; the acquisition terminal processes the received vibration signals in real time, extracts the amplitude values ​​of each order harmonic of the signal, and calculates the compaction evaluation index CEV value using the amplitude values ​​of each order harmonic of the signal, the road roller parameters, and the filler parameters to monitor the roadbed compaction quality. Chinese patent application number 201911214031, publication number CN110939040 A proposes a method and system for detecting the compaction quality of roadbed based on modal parameter identification. The method mainly includes: acquiring the position information and vibration signals during the compaction operation of a road roller, performing frequency domain analysis to obtain a frequency domain signal; identifying the natural frequency of the road roller-soil vibration system in the frequency domain signal using a modal parameter identification method; determining the change in stiffness of the road roller-soil vibration system by analyzing the change in its natural frequency, thereby achieving continuous and real-time compaction detection of the roadbed. Chinese patent application CN200910308318, publication number CN101672825 proposes a highway subgrade inspection system based on vibration spectrum analysis. The system mainly includes a detection device embedded in the subgrade pavement and a moving load analysis unit. The detection device detects vibration information of the subgrade pavement and transmits the detected vibration information to the moving load analysis unit via a wireless transmission unit. The moving load analysis unit applies a controllable frequency vibration force to the subgrade pavement and receives the vibration detection information transmitted by the detection device via a wireless receiving unit. It then performs frequency analysis on the received vibration detection information to provide information on the pavement quality at the location of the detection device embedded in the subgrade pavement.

[0005] The above continuous detection methods all utilize Fast Fourier Transform (FFT) to perform frequency domain analysis on acceleration signals, evaluating the subgrade compaction quality based on the ratio of harmonic to fundamental amplitudes in the Fourier spectrum. However, FFT assumes the signal is stationary, decomposing any signal into a weighted superposition of simple harmonic signals. Each simple harmonic signal corresponds to a fixed frequency and amplitude, failing to reflect the signal's temporal characteristics. Furthermore, the mechanism of harmonic occurrence in the Fourier spectrum is currently unclear. Field applications show that these frequency domain indicators are only suitable for fine-grained soil fillers in subgrades; their detection accuracy is low for coarse-grained soil fillers, failing to meet detection requirements. Therefore, there is an urgent need for an analytical method that accurately reflects the frequency domain characteristics of the vibratory wheel's dynamic response, and based on this, to propose continuous detection indicators for subgrade compaction quality, and establish corresponding detection and evaluation methods and systems. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention discloses a method for continuous detection and evaluation of roadbed compaction quality based on the frequency range of a vibrating wheel response. The technical solution is as follows:

[0007] A method for continuous detection and evaluation of roadbed compaction quality based on the frequency range of a vibrating wheel response, characterized by:

[0008] S1: The three-dimensional position information of the road roller during the compaction process is collected in real time by a satellite positioning receiver; the vertical acceleration signal of the vibrating wheel during the compaction process is collected in real time by an acceleration sensor; the three-dimensional position information of the road roller and the vertical acceleration signal of the vibrating wheel are transmitted to the on-board data processor via wired or wireless means;

[0009] S2: Segment the acceleration data based on satellite positioning information;

[0010] S3: The mode decomposition method is used to decompose each segment of the acceleration signal into a finite number of intrinsic mode functions (IMFs). Hilbert transform is performed on each IMF component to obtain the time-frequency-energy joint spectrum. Then, the acceleration energy spectrum is obtained by integrating over time.

[0011] S4: Calculate the response amplitude RV based on the acceleration energy spectrum;

[0012] S5: Based on the RV obtained from the continuous test of subgrade compaction quality and the conventional quality test indicators of the corresponding rolling area, establish a subgrade compaction quality test and evaluation model based on RV to evaluate the compaction quality of different areas of the subgrade surface.

[0013] This invention also discloses a continuous detection and evaluation system for roadbed compaction quality within the frequency range of a vibratory roller response. Its features include: a satellite positioning device, an acceleration acquisition device, and an onboard data processor; the satellite positioning receiver is installed on the top of the roller to acquire the roller's three-dimensional position information in real time during compaction; an acceleration sensor is installed on the roller's vibratory roller to acquire the roller's vertical acceleration signal in real time during compaction; the roller's position information and the roller's acceleration signal are transmitted to the onboard data processor via wired or wireless means; the onboard data processor processes the acceleration data based on the satellite positioning information. The system is segmented; the onboard data processor uses modal decomposition to decompose each segment of the acceleration signal into a finite number of intrinsic mode functions (IMFs). Hilbert transform is performed on each IMF component to obtain the time-frequency-energy joint spectrum, and then the acceleration energy spectrum is obtained by integrating over time. Based on the acceleration energy spectrum, the response amplitude RV is calculated. Based on the RV obtained from the continuous test of subgrade compaction quality and the conventional quality test indicators of the corresponding rolling area, a subgrade compaction quality test and evaluation model based on RV is established to evaluate the compaction quality of different areas of the subgrade surface.

[0014] Beneficial effects

[0015] 1. The RV index is obtained by calculating the energy spectrum through HHT processing of acceleration, which avoids the shortcomings of the current harmonic ratio index which uses FFT for spectrum analysis and assumes that the signal is stationary. It truly reflects the influence of subgrade compaction quality on the vibration wheel response amplitude.

[0016] 2. Field tests show that the correlation with conventional subgrade testing indicators (especially coarse-grained soil filler) is higher, and the testing accuracy is higher. Attached Figure Description

[0017] Figure 1 This is a continuous testing and evaluation process for roadbed compaction quality based on the vibration wheel response frequency;

[0018] Figure 2 RV calculation diagram;

[0019] Figure 3 The acceleration energy spectrum and RV of the packing material during the initial and later stages of compaction;

[0020] Figure 4 The relationship between RV and the dry density of roadbed fill material. Detailed Implementation

[0021] During vibratory compaction, the roller and the compacted fill material form a dynamic system, and the vertical dynamic response of the vibratory drum is related to the compaction state of the subgrade. Frequency domain characteristics are an important feature of the vibratory drum's dynamic response. The Hilbert-Huang Transform (HHT) can effectively handle nonlinear and non-stationary signals, accurately reflecting the frequency composition of the acceleration signal. Field test results show that as the fill material is compacted, the subgrade stiffness gradually increases, the nonlinearity of the vibratory drum acceleration signal intensifies, the frequency components of the acceleration change, and the vibration energy gradually disperses from the fundamental frequency to other frequencies. Based on this, this invention proposes a continuous detection index for subgrade compaction quality based on the frequency range of the vibratory drum response, and establishes a method and system for continuous quality detection and evaluation.

[0022] A method for continuous detection and evaluation of roadbed compaction quality based on the frequency range of vibration wheel response, comprising the following steps:

[0023] S1: The continuous monitoring system for roadbed compaction quality includes a satellite positioning device, an acceleration acquisition device, and an onboard data processor. The satellite positioning receiver is mounted on top of the road roller, acquiring its three-dimensional position information in real time during compaction. An acceleration sensor is mounted on the vibratory drum of the road roller, acquiring its vertical acceleration signal in real time during compaction. The road roller's position information and the vibratory drum's acceleration signal are transmitted to the onboard data processor via wired or wireless means. The acceleration sensor has a range of not less than ±10g and an acquisition frequency of not less than 1000Hz. The satellite positioning system has centimeter-level positioning capabilities. The onboard data processing system simultaneously stores the road roller's position information and acceleration signals, and processes the acceleration signals.

[0024] S2: The onboard data processor segments the acceleration data based on satellite positioning information. Acceleration data can be segmented into units based on a fixed distance (e.g., every 0.5m of roller travel) or a fixed time (e.g., every 1s of roller travel). To reduce spectral energy leakage, window functions (e.g., rectangular window, triangular window, Hanning window, Hamming window, Gaussian window, etc.) are used to truncate the signal. Each segment of acceleration data calculates and outputs a continuous detection value, which serves as the compaction quality detection value for the corresponding rolling area.

[0025] The calculation process of the window function:

[0026] 1. Preprocess the excitation signal w(t) and response signal λ(t), where t is time. The preprocessing method is polynomial least squares to eliminate the trend term. The order of the trend term elimination is 5. The excitation signal w5(t) and response signal λ5(t) after eliminating the trend term are obtained. The specific method is as follows:

[0027] 1) Perform least squares fitting on w(t) and y(t) to obtain the fitted curve h. x(t) and h y (t),;

[0028] 2) Eliminate the trend term, i.e., w1(t) = w(t) - h x (t), λ1(t) = λ(t) - h y (t);

[0029] 3) Repeat steps 1 and 2 for w1(t) and y1(t) 5 times to obtain u5(t) and λ5(t);

[0030] S3: The onboard data processor uses mode decomposition to decompose each segment of the acceleration signal into a finite number of intrinsic mode functions (IMFs). Hilbert transform is performed on each IMF component to obtain the time-frequency-energy joint spectrum, which is then integrated over time to obtain the acceleration energy spectrum. Empirical mode decomposition (EMD), ensemble empirical mode decomposition (EEMD), complementary ensemble empirical mode decomposition (CEEMD), and variational mode decomposition (VMD) methods are employed for mode decomposition of acceleration.

[0031] S4: Calculate the response amplitude RV based on the acceleration energy spectrum. After obtaining the acceleration energy spectrum, calculate RV.

[0032] The formula for calculating RA is as follows:

[0033] The time-history curve of the acceleration signal of the vibrating wheel is extracted and denoted as f(w). 1 / L The time-frequency-amplitude spectrum is then obtained by performing a Hibert-Huang transform on it. The specific calculation process is as follows:

[0034] 1) For f(t) 1 / L EMD (empirical mode decomposition) yields multiple IMFs (intrinsic mode functions):

[0035]

[0036] Where f(t) 1 / L The second part is the original acceleration signal extracted by the IMF. i (t) represents K intrinsic mode functions. Where r K The remaining term after subtracting the IMF from the signal.

[0037] 2) Given a signal f(t)1 / L, define the Hilbert transformation of the time-domain signal f(t)1 / L into a time-frequency-energy spectrum H[f(t)1 / L].

[0038]

[0039] Where t represents time, in seconds (s).

[0040] 3) The integral formula for the time-frequency-energy curve is:

[0041]

[0042] Where H(ω,t) is the Hilbert spectral function; e(ω,t) is the Hilbert marginal spectral function. W is the frequency;

[0043] 4) Integrating the calculated marginal spectrum yields the subgrade compaction index RV.

[0044] RV=Cf max

[0045] In the formula: C is a constant, and it is recommended to take a value between 1 and 20; f max This represents the maximum value of the energy spectrum.

[0046] Figure 3 The images show the vibrational acceleration energy spectra of the roller during the initial and later stages of subgrade fill compaction. It can be seen that in the initial stage of subgrade compaction, a peak value appears at the roller's vibration frequency, with a peak value of 0.3 g·s. In the later stage of compaction, the peak value increases to approximately 0.5–0.6 g·s. When C is 20, the RV values ​​in the initial and later stages of compaction are 6.44 and 14.26, respectively, with RV increasing significantly as the subgrade fill compacts. These phenomena indicate that as the fill is continuously compacted, the subgrade stiffness gradually increases, and the energy of the vibrational acceleration gradually increases, i.e., RV increases. Therefore, RV can be used for subgrade compaction quality detection and evaluation.

[0047] Figure 4 The results of two test strips conducted on a coarse-grained soil subgrade test section show that, with C = 20, the correlation R between RV and the dry density of the fill material is 0.95 and 0.97, respectively, which is much greater than the requirement of R > 0.7 in the current continuous compaction control specifications. This indicates that RV has high detection accuracy for coarse-grained soil fill material in subgrades.

[0048] S5: Based on the RV obtained from the continuous test of subgrade compaction quality and the conventional quality test indicators of the corresponding rolling area, establish a subgrade compaction quality test and evaluation model based on RV to evaluate the compaction quality of different areas of the subgrade surface.

[0049] After obtaining the conventional testing indicators of RV and its corresponding area, a univariate linear regression method was used to establish a subgrade compaction quality testing and evaluation model.

[0050] The methods and advantages of univariate regression:

[0051] The verification of continuous compaction indicators is conducted by establishing their correlation with conventional testing methods. Therefore, it directly reflects the correlation between continuous compaction quality indicators and conventional indicators. Thus, a univariate linear regression equation is used to test the correlation between the RV index and conventional indicators. The correlation is calculated according to the following formula:

[0052]

[0053] The correlation coefficient is calculated using the following formula:

[0054]

[0055] In the formula: x is the continuous compaction index; This represents the average value of continuous compaction indicators; y represents the value of a conventional indicator. x represents the average value of a common indicator; i With y i denoted as the i-th continuous compaction index and the conventional index value, respectively; n is the number of data points; a and b are regression coefficients; and i is a constant.

[0056] K = a0 + a1RV

[0057] Alternatively, simultaneously collect data such as RV (vehicle velocity), vehicle speed, vibration frequency, filler gradation, and filler moisture content, and establish a roadbed compaction quality testing and evaluation model based on RV using multiple linear regression, multiple nonlinear regression, or neural network methods.

[0058] Method for establishing a univariate regression model: The filler gradation, moisture content, vehicle speed, and vibration mode all affect RV. However, the univariate linear regression model only considers the compaction coefficient and does not consider the influence of the dispersion of filler parameters and rolling parameters, which may lead to prediction errors. Since the continuous detection of roadbed compaction quality is carried out in a weak vibration mode, this section first uses the compaction coefficient, vehicle speed, P5 content, and moisture content as independent variables of the multivariate regression model and RV as the dependent variable to establish a multivariate linear regression model as shown in equation (3). The regression results are shown in Table 2.

[0059] RV=a0+a1K+a2v+a3P5+a4ω (4)

[0060] In the formula: v is the vehicle speed, m / s2; P5 is the content of particles larger than 5mm in the packing, %; ω is the moisture content, %; a0 and a4 are regression coefficients.

[0061] Results of linear regression significance analysis

[0062]

[0063] As shown in Table 2, the significance levels for compaction coefficient, vehicle speed, and P5 content are all less than 0.050, indicating a clear linear correlation between these three factors and CEV. However, the significance level for moisture content is 0.883, greater than 0.050, indicating that for this experiment, the impact of moisture content on CEV is far lower than that of compaction coefficient, vehicle speed, and P5 content. According to the experimental results in Section 2.4, the impact of moisture content on CEV is relatively small when it is within the optimum moisture content range of -3% to 2%. In this experiment, the moisture content range of the 32 measuring points in the two test strips was 4.53% to 7.00%, with small variations and all within the optimum moisture content range; therefore, its impact on CEV is not significant. These results indicate that during subgrade compaction construction, under strict moisture content control, its impact on continuous compaction quality monitoring can be ignored. Therefore, moisture content can be disregarded in modeling, and the multiple regression model can be further constructed as follows:

[0064] CEV=a0+a1K+a2v+a3P5 (5)

[0065] The symbols in the formula have the same meaning as in formula (4). The results of the repeated multivariate regression analysis are shown in Table 3.

[0066] Table 3. Results of linear regression significance analysis

[0067]

[0068] K = f(RV, v, f, P5, ω, ...)

[0069] In the formula: K represents the conventional test index of roadbed compaction quality, including compaction coefficient and dynamic deformation modulus E. vd subgrade coefficient K 30 etc.; v is the speed of the road roller; f is the vibration frequency; P5 is the content of particles larger than 5mm; ω is the moisture content.

[0070] This invention performs frequency domain processing on the vibration wheel acceleration signal using HHT transform, accurately obtaining the frequency components and their corresponding energy amplitudes. By analyzing the relationship between the vibration wheel response frequency range RV and conventional testing indicators, a continuous testing and evaluation method and system for roadbed compaction quality is established. Compared with traditional roadbed compaction quality testing methods, this invention enables real-time, continuous, accurate, and comprehensive testing of roadbed surface compaction quality. Compared with existing continuous roadbed compaction quality testing methods, RV is calculated through the acceleration energy spectrum amplitude, avoiding the disadvantage of assuming a stationary signal in harmonic ratio-type indicators using FFT for spectrum analysis. It has a higher correlation with conventional roadbed testing indicators (especially coarse-grained soil fillers), resulting in higher testing accuracy. It is widely applicable to the compaction quality testing of various filler types in railway and highway roadbeds, and can also be extended to water conservancy and airport projects.

[0071] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for continuously detecting and evaluating the compaction quality of a roadbed based on the frequency range of a vibrating wheel response, characterized by: S1: collecting real-time three-dimensional position information of a road roller during the rolling process through a satellite positioning receiver; collecting real-time vertical acceleration signals of the vibrating wheel during the rolling process through an acceleration sensor; and transmitting the three-dimensional position information of the road roller and the vertical acceleration signals of the vibrating wheel to a vehicle-mounted data processor through wired or wireless means; S2: segmenting the acceleration data according to the satellite positioning information; S3: decomposing each segmented acceleration signal into a finite number of intrinsic mode functions (IMFs) using a modal decomposition method, performing Hilbert transformation on each order of IMF component to obtain a time-frequency-energy joint spectrum, and then integrating the time to obtain an acceleration energy spectrum; S4: calculating a response amplitude (RV) based on the acceleration energy spectrum; S5: establishing a roadbed compaction quality detection evaluation model based on the RV obtained from the continuous detection test of the roadbed compaction quality and the conventional quality detection indicators of the corresponding rolling area, and evaluating the compaction quality of different areas of the roadbed surface; the acceleration sensor has a range of not less than ±10g and a collection frequency of not less than 1000Hz; the vehicle-mounted data processor synchronously stores the position information of the road roller and the acceleration signals, and processes the acceleration signals; the acceleration data is segmented according to a fixed distance or a fixed time as a unit; a window function is used to truncate the signal, and each segmented acceleration data calculates and outputs a continuous detection value as the compaction quality detection value of the corresponding rolling area of the unit acceleration; the acceleration is modal decomposed using the empirical mode decomposition (EMD), ensemble empirical mode decomposition (EEMD), or complementary ensemble empirical mode decomposition (CEEMD), or variational mode decomposition (VMD) method; in step S4, after obtaining the acceleration energy spectrum, the RV is calculated, and the RV is calculated according to the following formula: The system comprises a satellite positioning device, an acceleration collection device, and a vehicle-mounted data processor; the satellite positioning receiver is installed on the top of the road roller, and collects real-time three-dimensional position information of the road roller during the rolling process; the acceleration sensor is installed on the vibrating wheel of the road roller, and collects real-time vertical acceleration signals of the vibrating wheel during the rolling process; the position information of the road roller and the acceleration signals of the vibrating wheel are transmitted to the vehicle-mounted data processor through wired or wireless means; The vehicle-mounted data processor segments the acceleration data according to the satellite positioning information; The vehicle-mounted data processor decomposes each segmented acceleration signal into a finite number of intrinsic mode functions (IMFs) using a modal decomposition method, performs Hilbert transformation on each order of IMF component to obtain a time-frequency-energy joint spectrum, and then integrates the time to obtain an acceleration energy spectrum; calculates a response amplitude (RV) based on the acceleration energy spectrum; and establishes a roadbed compaction quality detection evaluation model based on the RV obtained from the continuous detection test of the roadbed compaction quality and the conventional quality detection indicators of the corresponding rolling area, and evaluates the compaction quality of different areas of the roadbed surface. ​ ​ RV = Cf max In the formula, C is a constant, taking 1-20; f max is the maximum value of the energy spectrum.

2. A system for continuous evaluation of subgrade compaction quality in response to the frequency range of a vibratory roller, the evaluation system comprising the evaluation method of claim 1, characterized in that: ​ ​ ​ 3. A non-volatile storage medium, characterized by, The non-volatile storage medium includes a stored program, wherein the program, when executed, controls the device where the non-volatile storage medium resides to perform the method described in claim 1.

Images