A road surface intelligent adaptive detection and maintenance method and system

Abstract

The present invention discloses a method and system for intelligent adaptive detection and maintenance of road surface. Multiple sensors monitor traffic and road conditions in real time to obtain amplitude and frequency data of road surface vibration. The detected road surface information status data is used as input, and the AVOA algorithm is used to optimize the parameters of the VMD decomposition module using the above objective function. The optimized VMD decomposition module is used to decompose and process the road surface status, and the road surface detection result is obtained according to the decomposed high-frequency signal, and the road surface quality maintenance plan is further determined. Compared with the prior art, the present invention can better adapt to the characteristics of road surface vibration signals and improve the accuracy and stability of detection by optimizing the parameters in the VMD algorithm using the AVOA algorithm. The VMD decomposition module is used to decompose the signal, and the quality of the road surface is further analyzed by analyzing the high-frequency signal. Targeted measures can be taken according to needs to carry out repairs and reinforcements.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent algorithms and road surface state prediction, and in particular to a road surface intelligent adaptive detection and maintenance method and system. Background Art

[0002] In recent years, China's modernization construction has developed rapidly. Transportation is a pillar supporting economic development. Highways, as the main traditional mode of transportation, have also been greatly developed. However, there are some problems and shortcomings in the development of highways. Highway maintenance is one of the problems at this stage. Due to the rapid development of the economy, more and more people have cars, and more and more people go out to play. There are many problems such as damage on the highway. At present, the road surface inspection adopts regular inspection, generally using manual visual inspection to regularly inspect the road surface, including cracks, potholes, settlement and other damage monitoring, and the maintenance method is also manual repair.

[0003] Therefore, how to realize the intelligent inspection and maintenance of the road surface is an issue that needs to be considered urgently. The present invention proposes an intelligent inspection and maintenance technology. Since the main characteristics of non-destructive automatic inspection are rapidity and intelligence, multi-source sensors are used to work together and are integrated on a multi-functional road inspection vehicle. The amplitude and frequency data of road vibration can be detected and signal processing can be realized through VMD decomposition. The road surface damage, flatness, rutting, anti-skid performance and structural strength as well as the road line shape and facilities along the road can be determined according to the decomposed signals. Summary of the invention

[0004] Purpose of the invention: In order to overcome the deficiencies of the prior art, the present invention provides a method and system for intelligent adaptive detection and maintenance of road surfaces, which solves the road surface detection problem in the prior art and clearly obtains road surface status information.

[0005] Technical solution: The present invention discloses a road surface intelligent adaptive detection and maintenance method, comprising the following steps:

[0006] Step 1: Use sensors to collect real-time status information of the road surface on which the vehicle passes, obtain the amplitude and frequency data of road vibration, and determine the crack width, crack density, pothole depth, and damaged area of ​​the road surface;

[0007] Step 2: Construct the following objective function:

[0008]

[0009]

[0010]

[0011] Among them, d iThe indicator representing the location of the road surface fault is assumed to be a vibration signal sequence x = [x1, x2, ..., x N ], where x i represents the vibration signal value of the i-th sampling point, and defines an abnormality measurement index D = [d1, d2, ..., d N ], indicating the abnormal vibration degree of the road surface at each position; y i The sensor collects the amplitude and frequency data of road vibration; is the value decomposed by variational mode decomposition VMD; N is the number of samples; ω1 and ω2 are weight coefficients to balance the influence of the two indicators; O is the objective function of road quality detection accuracy;

[0012] Step 3: Using the detected road surface information state data as input, the AVOA algorithm is used to optimize the parameters γ = (K, α) of the VMD decomposition module with the above objective function;

[0013] Step 4: Use the optimized VMD decomposition module to decompose the road pavement status, obtain the pavement detection results based on the decomposed high-frequency signals, and further determine the pavement quality maintenance plan.

[0014] Furthermore, the real-time status information of the vehicle passing through the road surface in step 1 is collected by means of sensors installed on the road surface or on the vehicle. The sensors directly measure the vibration of the vehicle and record in real time the vibration amplitude and frequency of the vehicle when it travels on different road surfaces.

[0015] Furthermore, the specific steps of step 3 using the AVOA algorithm to optimize the parameters of the VMD decomposition module with the above objective function are as follows:

[0016] (1) Determine the parameter space: input the detected road surface information status data, set the range of the VMD decomposition module parameter γ = (K, α) to be optimized, and initialize the AVOA parameters, including the maximum number of iterations L and the number of search groups n;

[0017] (2) Initialize AVOA parameters: After the optimal vulture initial population is formed, calculate the fitness values ​​of all solutions in the population, and select the optimal solution as the optimal vulture of the first group, and the suboptimal solution as the optimal vulture of the second group. Other vultures move toward the optimal and suboptimal vultures according to the following formula:

[0018]

[0019] Where: L1 and L2 are original parameters, whose values ​​are between 0 and 1, and their sum is 1; p i is the probability of selecting the best vulture;

[0020] (3) Randomly initialize the positions of African vultures, calculate the objective function of all African vultures, that is, the accuracy of road quality detection, and save the largest parameter combination γ = (K, α), as follows:

[0021] If a vulture is well-fed, it has higher energy and can forage in a wider area; if it is hungry, it can only forage next to strong vultures. The hunger rate is expressed as:

[0022]

[0023]

[0024] Where: F is the vulture hunger rate; Z is a random number between -1 and 1; Iterationi is the current number of iterations; maxiterations is the maximum number of iterations; h is a random number between -2 and 2; rand1 is a random number between 0 and 1; ω is a control parameter, which controls the probability of the algorithm entering the exploration phase at the end;

[0025] (4) Iterative optimization: Through the iterative process, the AVOA algorithm is used to optimize the individuals in the population. In each generation, the performance of the individuals is evaluated according to the fitness function, and the characteristics of the adaptive variables are used to adjust the parameters to find a better parameter setting. According to the AVOA algorithm, the position of the African vulture population is updated, the objective function of the African vulture, i.e., the accuracy of road quality detection, is recalculated, and the optimal solution is updated. The details are as follows:

[0026] Vultures have strong visual and predation abilities. In AVOA, the following formula is used to explore whether different random areas are suitable for predation:

[0027]

[0028] D(i)=|X×J(i)-P(i)|

[0029] Where: P(i+1) is the updated position of the vulture in the next iteration; O(i) is the best vulture selected; ub, lb are the upper and lower limits of the search space; rand2, rand3, randp1 are random numbers between 0 and 1; X is the position where the vulture moves randomly; P(i) is the current vulture position;

[0030] (5) Termination condition: At the time of termination, the individual with the best fitness is selected as the optimal parameter setting, and it is judged whether the termination condition is met. If it is met, it jumps out to find the optimal solution for the accuracy of the road quality detection; otherwise, it repeats step (3); the specific steps are as follows:

[0031] When the absolute value of the hunger rate is 0.5≤|F|≤1, AVOA enters the first stage of the development phase:

[0032]

[0033] When the absolute value of the starvation rate |F| < 0.5, AVOA enters the second stage of the development phase:

[0034]

[0035] Where: BestV1(t) and BestV2(t) are the positions of the optimal vulture and the sub-optimal vulture in the t-th iteration; rand4, rand5, rand6, randp2, randp3 are random numbers between 0 and 1; Levy(d) is a random walk, which can improve the computational efficiency of AVOA;

[0036] (6) Apply the optimized optimal parameter optimal combination γ = (K, α) to the VMD decomposition module.

[0037] Furthermore, the specific steps for the optimized VMD decomposition module in step 4 to decompose the road pavement state are as follows:

[0038] Perform initialization on VMD λ 1 :

[0039]

[0040] Where: λ is the Lagrange operator; α is the quadratic penalty factor. The penalty factor determines the bandwidth of the IMF components. The smaller the penalty coefficient, the larger the bandwidth of each IMF component; the larger the β value, the smaller the bandwidth of each IMF component. The IMF components reflect the information characteristics of the image from high frequency to low frequency;

[0041] Update the parameters u k , ω k : Update the parameter λ;

[0042]

[0043]

[0044]

[0045] When the given accuracy is met, the iteration terminates, as shown in the following formula:

[0046]

[0047] Judge the convergence according to the termination condition formula. If the condition is met, terminate the iteration; if the condition is not met and n < N, repeat the above steps.

[0048] Furthermore, the collected vibration signal is decomposed through the VMD decomposition module to extract the various frequency components in the signal, and the signal is decomposed into band-limited intrinsic mode functions IMFs. Each IMF reflects a specific frequency range in the signal. The high-frequency IMFs obtained after VMD decomposition are checked. If a sudden increase in the vibration amplitude or a significant change in the frequency component is found in the high-frequency IMFs, it is judged that the road surface is abnormal.

[0049] The present invention also discloses a system based on the above-mentioned road surface intelligent adaptive detection and maintenance method, including a road surface quality detection module, an AVOA algorithm optimization module, a VMD decomposition module and a road surface quality maintenance module;

[0050] A road quality detection module is used to collect the amplitude and frequency data of road vibration through sensors installed on the road surface or on the vehicle, and determine the crack width, crack density, pothole depth and damaged area;

[0051] The AVOA algorithm optimization module is used to take the detected road surface information status data as input, use the AVOA algorithm to maximize the road surface quality detection accuracy as the optimization goal, and optimize the parameters of the VMD decomposition module;

[0052] The VMD decomposition module is used to process the collected signal through variational mode decomposition using the optimized VMD decomposition module, decomposing it into multiple modal functions IMFs with different frequencies, and further obtaining the road surface detection results;

[0053] The pavement quality maintenance module is used to identify pavement anomalies through high-frequency IMFs, so that repair or maintenance strategies can be formulated and implemented according to the nature of the anomalies.

[0054] Beneficial effects:

[0055] The present invention proposes an intelligent adaptive maintenance technology for a road surface. The amplitude and frequency data of road surface vibration are detected by road surface quality detection technology. The high frequency and low frequency are decomposed into the improved VMD decomposition module to analyze the road surface data and obtain the state of the road surface. The present invention uses the AVOA algorithm to perform parameter optimization to improve the accuracy of the VMD decomposition algorithm in road surface quality detection. By optimizing the parameters in the VMD algorithm, such as the number of modes and the number of frequency bands, the algorithm can better adapt to the characteristics of the road surface vibration signal and improve the accuracy and stability of the detection. Determine the parameter space: During the optimization process, it is necessary to determine the parameter space, including the relevant parameter range in the VMD algorithm, such as the number of modes and the number of frequency bands. By reasonably setting the parameter space, the optimal parameter combination can be searched more effectively. Optimization effect: By optimizing the VMD algorithm parameters through the AVOA algorithm, the accuracy and stability of road surface quality detection can be improved, the detection results can be more reliable and accurate, and it is helpful to timely discover and solve road surface quality problems.

[0056] In summary, VMD pavement quality detection combined with parameter optimization of AVOA algorithm can improve detection accuracy, provide more reliable technical support for pavement quality assessment, help maintain and improve road quality, and enhance traffic safety and efficiency. The development and application of these technologies will have a positive impact on the field of road construction and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a system structure framework diagram of the present invention;

[0058] Figure 2 The improved VMD algorithm flow chart is adopted in the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] like Figure 1 As shown, the present invention discloses a road surface intelligent adaptive detection and maintenance method and system, including: a road surface quality detection module, an AVOA algorithm optimization module, a VMD decomposition module and a road surface quality maintenance module.

[0061] A road surface intelligent adaptive detection and maintenance method comprises the following steps:

[0062] Step 1: The road quality detection module collects real-time status information of vehicles passing through the road surface through sensors, obtains the amplitude and frequency data of road vibration, determines the crack width, crack density, pothole depth and damaged area of ​​the road surface, and collects real-time status information of vehicles passing through the road surface through sensors installed on the road surface or on the vehicle. The sensor directly measures the vibration of the vehicle and records the vibration amplitude and frequency of the vehicle in real time when it is driving on different road surfaces.

[0063] The sensors installed on the road surface are used to monitor the road conditions in real time. The sensors installed on the vehicle can directly measure the vibration of the vehicle. These sensors (such as accelerometers, pressure sensors, GPS positioning, geomagnetic sensors) can record the vibration amplitude and frequency of the vehicle when it is driving on different roads in real time. By analyzing these data, the road conditions can be inferred. Data collection can be performed during the normal driving of the vehicle without causing interference to traffic. The data collected by the on-board sensors can be uploaded to the cloud platform for summary and analysis. The road conditions are obtained through the VMD analysis method and uploaded to the cloud platform.

[0064] Step 2: Construct the following objective function:

[0065]

[0066]

[0067]

[0068] Among them, d i The indicator representing the location of the road surface fault is assumed to be a vibration signal sequence x = [x1, x2, ..., x N ], where x i represents the vibration signal value of the i-th sampling point, and defines an abnormality measurement index D = [d1, d2, ..., d N ], indicating the abnormal vibration degree of the road surface at each position; y i The sensor collects the amplitude and frequency data of road vibration; is the value decomposed by variational mode decomposition VMD; N is the number of samples; ω1 and ω2 are weight coefficients to balance the influence of the two indicators; O is the objective function of road quality detection accuracy;

[0069] Step 3: The AVOA algorithm optimization module takes the detected road surface information state data as input, and uses the AVOA algorithm to optimize the parameters γ = (K, α) of the VMD decomposition module with the above objective function;

[0070] Step 4: The VMD decomposition module uses the optimized VMD decomposition module to decompose the road pavement status, obtain the pavement detection results based on the decomposed high-frequency signals, and further determine the pavement quality maintenance plan.

[0071] The VMD decomposition module is used to process the collected signal through variational mode decomposition and decompose it into multiple modal functions IMFs of different frequencies. VMD can adaptively determine the number of decompositions and the frequency bandwidth, so as to accurately extract high-frequency and low-frequency components from complex signals, so as to further analyze the quality of the road surface. Analyze high-frequency IMFs and look for sharp changes or unstable signals in the vibration signal, which usually indicates that there are abnormalities in the road surface, such as potholes, cracks or other surface damage. Low-frequency IMFs reflect the overall condition or long-term change trend of the road surface, which is useful for evaluating the overall health of the road surface and predicting future maintenance needs.

[0072] The VMD decomposition module is used to decompose and process the signal taken from the road quality detection module. The VMD algorithm first decomposes the input road vibration signal to obtain a series of IMFs, each of which corresponds to information in a different frequency range. Vibration signals usually refer to vibration signals generated when a vehicle passes through the road surface. When a vehicle travels on the road surface, the contact between the vehicle's tires and the road surface will cause vibration, which can reflect key characteristics such as the flatness and damage of the road surface. These IMFs contain high-frequency and low-frequency components of the signal, which help to reveal the vibration characteristics and noise of the road surface. By analyzing these frequency components, the vibration characteristics, noise conditions and other related characteristics of the road surface can be extracted. These characteristics can include information such as the distribution of frequency components, amplitude changes, and energy density, which helps to understand the actual situation of the road surface. Based on the extracted features, the quality of the road surface can be further analyzed. Abnormal signals or high-frequency signals may indicate damage or unevenness of the road surface, while low-frequency signals can provide information about the overall condition of the road surface. According to the input time and distance information, it can be traced back to the specific road section range, thereby realizing the accurate positioning and tracking of road quality problems. This helps to take maintenance measures in a timely manner to ensure the safety and reliability of the road.

[0073] Road surface abnormality judgment criteria: Sudden increase in vibration amplitude Compared with the vibration level under normal driving conditions, a significant increase in vibration amplitude usually indicates that there is an abnormality in the road surface; Frequency change The frequency of the vibration signal in the abnormal road surface area will be significantly different from that in the normal road surface area, especially in the high-frequency range; Energy is concentrated in the frequency domain. The vibration caused by the abnormality often causes the energy to be concentrated in certain frequency ranges, which is different from the energy distribution under normal road conditions.

[0074] The pavement quality maintenance module is used to quickly develop and implement repair or maintenance strategies based on the nature of the abnormality once the pavement anomaly is identified through high-frequency IMFs. For example, small potholes detected can be partially filled; cracks found can be encapsulated. By promptly identifying problems and performing maintenance, the service life of the pavement can be extended and driving comfort and safety can be improved.

[0075] The specific steps of using the AVOA algorithm to optimize the parameters of the VMD decomposition module with the above objective function are as follows:

[0076] (1) Determine the parameter space: Input the detected road surface information status data, set the range of the VMD decomposition module parameter γ = (K, α) to be optimized, and initialize the AVOA parameters, including the maximum number of iterations L and the number of search groups n.

[0077] (2) Initialize AVOA parameters: After the optimal vulture initial population is formed, calculate the fitness values ​​of all solutions in the population, and select the optimal solution as the optimal vulture of the first group, and the suboptimal solution as the optimal vulture of the second group. Other vultures move toward the optimal and suboptimal vultures according to the following formula:

[0078]

[0079] Where: L1 and L2 are original parameters, whose values ​​are between 0 and 1, and their sum is 1; p i is the probability of selecting the best vulture.

[0080] (3) Randomly initialize the positions of African vultures, calculate the objective function of all African vultures, that is, the accuracy of road quality detection, and save the largest parameter combination γ = (K, α), as follows:

[0081] If a vulture is well-fed, it has higher energy and can forage in a wider area; if it is hungry, it can only forage next to strong vultures. The hunger rate is expressed as:

[0082]

[0083]

[0084] Where: F is the vulture hunger rate; Z is a random number between -1 and 1; Iterationi is the current number of iterations; maxiterations is the maximum number of iterations; h is a random number between -2 and 2; rand1 is a random number between 0 and 1; ω is a control parameter, which controls the probability of the algorithm entering the exploration stage at the end.

[0085] (4) Iterative optimization: Through the iterative process, the AVOA algorithm is used to optimize the individuals in the population. In each generation, the performance of the individuals is evaluated according to the fitness function, and the characteristics of the adaptive variables are used to adjust the parameters to find a better parameter setting; according to the AVOA algorithm, the position of the African vulture population is updated, the objective function of the African vulture, that is, the accuracy of the road quality detection, is recalculated, and the optimal solution is updated. The details are as follows:

[0086] Vultures have strong visual and predation abilities. In AVOA, the following formula is used to explore whether different random areas are suitable for predation:

[0087]

[0088] D(i)=|X×J(i)-P(i)|

[0089] Where: P(i+1) is the updated position of the vulture in the next iteration; O(i) is the best vulture selected; ub, lb are the upper and lower limits of the search space; rand2, rand3, randp1 are random numbers between 0 and 1; X is the position where the vulture moves randomly; P(i) is the current vulture position.

[0090] (5) Termination condition: At the time of termination, the individual with the best fitness is selected as the optimal parameter setting to determine whether the termination condition is met. If so, the process will exit to obtain the optimal solution for the accuracy of the road quality detection; otherwise, the process will repeat step (3). The specific steps are as follows:

[0091] When the absolute value of the hunger rate is 0.5≤|F|≤1, AVOA enters the first stage of the development phase:

[0092]

[0093] When the absolute value of the hunger rate |F| < 0.5, AVOA enters the second stage of the development phase:

[0094]

[0095] Where: BestV1(t), BestV2(t) are the positions of the best vulture and the second best vulture in the tth iteration; rand4, rand5, rand6, randp2, randp3 are random numbers between 0 and 1; Levy(d) is a random walk, which can improve the calculation efficiency of AVOA.

[0096] (6) Apply the optimal combination of optimized parameters γ = (K, α) to the VMD decomposition module, and apply the optimized optimal parameters to the VMD pavement quality detection algorithm to improve the accuracy of detection. Through the above steps, combined with the characteristics of the AVOA algorithm for parameter optimization, the accuracy of VMD pavement quality detection can be effectively improved, making the detection system more robust and reliable.

[0097] The optimized VMD decomposition module performs the following steps to decompose the road surface conditions:

[0098] VMD λ 1 To initialize:

[0099]

[0100] Where: λ is the Lagrange operator; α is the quadratic penalty factor, which determines the bandwidth of the IMF component. The smaller the penalty coefficient, the larger the bandwidth of each IMF component; the larger the β value, the smaller the bandwidth of each IMF component. The IMF component reflects the information characteristics of the image from high frequency to low frequency.

[0101] For parameter u k , ω k Update: Update the parameter λ;

[0102]

[0103]

[0104]

[0105] When the given accuracy is met, the iteration terminates, as shown in the following formula:

[0106]

[0107] Judge the convergence according to the termination condition formula. If the condition is met, terminate the iteration; if the condition is not met and n < N, repeat the above steps.

[0108] Based on the adaptive VMD method with O as the objective function, parameter optimization is achieved through the AVOA algorithm to optimize the accuracy of pavement quality detection for the purpose of better detecting the pavement quality.

[0109] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable those who are familiar with this technology to understand the content of the present invention and implement it accordingly, and should not be used to limit the protection scope of the present invention. Any equivalent transformation or modification made according to the spirit and essence of the present invention should be covered within the protection scope of the present invention.

Claims

1. A road surface intelligent adaptive detection and maintenance method, characterized in that: It includes the following steps: Step 1: Collect the real-time state information of the vehicle passing through the road surface through sensors, obtain the amplitude and frequency data of the road surface vibration, and determine the crack width, crack density, pothole depth, and damage area of the road surface; Step 2: Construct the following objective function: Among them, d i The indicator representing the location of the road surface fault is assumed to be a vibration signal sequence x = [x1, x2, ..., x N ], where x i represents the vibration signal value of the i-th sampling point, and defines an abnormality measurement index D = [d1, d2, ..., d N ], indicating the abnormal vibration degree of the road surface at each position; y i The sensor collects the amplitude and frequency data of road vibration; is the value decomposed by variational mode decomposition VMD; N is the number of samples; ω1 and ω2 are weight coefficients to balance the influence of the two indicators; O is the objective function of road quality detection accuracy; Step 3: Using the detected road surface information state data as input, utilize the AVOA algorithm to optimize the parameter γ=(K,α) of the VMD decomposition module with the above objective function; Step 4: Use the optimized VMD decomposition module to decompose the road surface state, obtain the road surface detection result according to the decomposed high-frequency signal, and further determine the road surface quality maintenance plan; right λ 1 To initialize: In the formula: λ is the Lagrange operator; α is the quadratic penalty factor, and the penalty factor determines the bandwidth of the IMF component. The smaller the penalty coefficient, the larger the bandwidth of each IMF component; the larger the β value, the smaller the bandwidth of each IMF component, and the IMF component reflects the information characteristics of the image from high frequency to low frequency; For parameter u k ,ω k Update: Update the parameter λ; When the given accuracy is met, the iteration terminates, as shown in the following formula: Judge the convergence according to the termination condition formula. If the condition is met, terminate the iteration; if the condition is not met and n < N, repeat the above steps; Decompose the collected vibration signal through the VMD decomposition module to extract each frequency component in the signal, decompose the signal into band-limited intrinsic mode functions IMFs, and each IMF reflects a specific frequency range in the signal. Check the high-frequency IMFs obtained after VMD decomposition. If a sudden increase in vibration amplitude or a significant change in frequency components is found in the high-frequency IMFs, it is judged as a road surface anomaly.

2. The method for intelligent adaptive road surface detection and maintenance according to claim 1, characterized in that: In the said Step 1, collecting the real-time state information of the vehicle passing through the road surface is realized by sensors installed on the road surface or the vehicle. The sensors directly measure the vibration of the vehicle and record the vibration amplitude and frequency of the vehicle when driving on different road surfaces in real time.

3. The method for intelligent adaptive road surface detection and maintenance according to claim 1, characterized in that: The specific steps of using the AVOA algorithm in the said Step 3 to optimize the parameter of the VMD decomposition module with the above objective function are as follows: (1) Determine the parameter space: Input the detected road surface information state data, set the range of the parameter γ=(K,α) of the VMD decomposition module to be optimized, and initialize the AVOA parameters, including the maximum number of iterations L and the number of search groups n; (2) Initialize the AVOA parameters: After determining the formation of the optimal vulture initial population, calculate the fitness values of all solutions in the population, and select the optimal solution as the optimal vulture of the first group and the sub-optimal solution as the optimal vulture of the second group. Other vultures move towards the optimal and sub-optimal vultures through the following formula: Where: L1 and L2 are original parameters, whose values ​​are between 0 and 1, and their sum is 1; p i is the probability of selecting the best vulture; (3) Randomly initialize the positions of the African vultures, calculate the objective function of all African vultures, that is, the accuracy of road surface quality detection, and save the maximum parameter combination γ=(K,α), specifically as follows: If the vulture is in a satiated state, it has higher energy and can thus forage in a wider area; if it is in a hungry state, it can only forage beside strong vultures. The hunger rate is expressed as: Where: F is the vulture hunger rate; Z is a random number between -1 and 1; Iterationi is the current number of iterations; maxiterations is the maximum number of iterations; h is a random number between -2 and 2; rand1 is a random number between 0 and 1; ω is a control parameter, which controls the probability of the algorithm entering the exploration phase at the end; (4) Iterative optimization: Through the iterative process, the AVOA algorithm is used to optimize the individuals in the population. In each generation, the performance of the individuals is evaluated according to the fitness function, and the characteristics of the adaptive variables are used to adjust the parameters to find a better parameter setting. According to the AVOA algorithm, the position of the African vulture population is updated, the objective function of the African vulture, i.e., the accuracy of road quality detection, is recalculated, and the optimal solution is updated. The details are as follows: Vultures have strong visual and predation abilities. In AVOA, the following formula is used to explore whether different random areas are suitable for predation: D(i)=|X×J(i)-P(i)| Where: P(i+1) is the updated position of the vulture in the next iteration; O(i) is the best vulture selected; ub, lb are the upper and lower limits of the search space; rand2, rand3, randp1 are random numbers between 0 and 1; X is the position where the vulture moves randomly; P(i) is the current vulture position; (5) Termination condition: At the time of termination, the individual with the best fitness is selected as the optimal parameter setting, and it is judged whether the termination condition is met. If it is met, it jumps out to find the optimal solution for the accuracy of the road quality detection; otherwise, it repeats step (3); the specific steps are as follows: When the absolute value of the hunger rate is 0.5≤|F|≤1, AVOA enters the first stage of the development phase: When the absolute value of the hunger rate |F| < 0.5, AVOA enters the second stage of the development phase: Where: BestV1(t), BestV2(t) are the positions of the best vulture and the second best vulture in the tth iteration; rand4, rand5, rand6, randp2, randp3 are random numbers between 0 and 1; Levy(d) is a random walk, which can improve the calculation efficiency of AVOA; (6) The optimal combination of optimized parameters γ = (K, α) is applied to the VMD decomposition module.

4. A system based on the road surface intelligent adaptive detection and maintenance method according to any one of claims 1 to 3, characterized in that: It includes pavement quality detection module, AVOA algorithm optimization module, VMD decomposition module and pavement quality maintenance module; A road quality detection module is used to collect the amplitude and frequency data of road vibration through sensors installed on the road surface or on the vehicle, and determine the crack width, crack density, pothole depth and damaged area; The AVOA algorithm optimization module is used to take the detected road surface information status data as input, use the AVOA algorithm to maximize the road surface quality detection accuracy as the optimization goal, and optimize the parameters of the VMD decomposition module; The VMD decomposition module is used to process the collected signal through variational mode decomposition using the optimized VMD decomposition module, decomposing it into multiple modal functions IMFs with different frequencies, and further obtaining the road surface detection results; The pavement quality maintenance module is used to identify pavement anomalies through high-frequency IMFs, so that repair or maintenance strategies can be formulated and implemented according to the nature of the anomalies.

Images