Low-traffic-volume road grading detection evaluation method and device based on pavement performance prediction

Abstract

The invention discloses a low-traffic-volume road grading detection evaluation method and device based on pavement performance prediction, and the method comprises the steps: constructing an annual average daily traffic volume condition inference tree of a to-be-evaluated region, and obtaining a traffic volume grouping boundary; performing road network screening on the to-be-evaluated region based on the traffic volume grouping boundary, and selecting target road sections of different traffic volume groups; constructing a PPI attenuation curve based on the composite pavement performance decay function, and obtaining a maintenance time window based on the PPI attenuation curve; based on the maintenance time window, constructing an objective function about the comprehensive detection cost of the operation period, and adopting Weibull distribution considering the reliability to restrain the probability of missing detection risk in the objective function; and on the basis of the objective function and the traffic volume classification, determining recommended regular check periods of different traffic volume groups, thereby solving the problem that a classification detection mode for low-traffic-volume roads in the traditional technology cannot form a complete comprehensive evaluation system.

Description

Technical Field

[0001] This application relates to the field of road engineering technology, and more specifically, to a method and apparatus for classifying and evaluating low-traffic roads based on pavement performance prediction. Background Technology

[0002] With the vigorous construction of national transportation infrastructure and rapid urbanization, a trunk highway network with expressways as its backbone has been basically formed. However, regional development differences have led to significant variations in the average daily traffic volume of highways of the same grade within the same network. Current road maintenance standards do not fully consider the nonlinear impact of traffic volume dispersion on the rate of pavement performance degradation, resulting in high redundancy in the frequency of inspections on low-traffic sections and a surge in human and material costs. Furthermore, the lack of a correlation model between traffic volume classification and the timing of preventative maintenance makes it difficult for road operators to gain a holistic understanding of the pavement performance status of low-traffic roads, hindering the ability to meet the needs of tiered decision-making. Summary of the Invention

[0003] This invention provides a method and apparatus for classifying and evaluating low-traffic roads based on pavement performance prediction, to solve the technical problems of the prior art as described in the background section. The method includes: Construct a conditional inference tree for the annual average daily traffic volume of the region to be evaluated to obtain the traffic volume grouping boundaries; Based on the traffic volume grouping boundaries, road network screening is performed on the area to be evaluated, and target road segments with different traffic volume groups are selected. A PPI decay curve is constructed based on a composite pavement performance decay function, and a maintenance time window is obtained based on the PPI decay curve. Based on the maintenance time window, an objective function for the comprehensive inspection cost during the operation period is constructed, and a reliability-considering approach is adopted. The Weibull distribution constrains the probability of missed detection in the objective function; Based on the objective function and the traffic volume classification, the recommended maintenance cycle for different traffic volume groups is determined.

[0004] In some specific embodiments, an annual average daily traffic volume condition inference tree for the region to be evaluated is constructed to obtain traffic volume grouping boundaries, specifically as follows: A decision tree is constructed based on the conditional inference framework, and the log-standardized traffic volume is selected. As a continuous predictor variable, the trend of pavement performance change is selected as the response variable, and the trend is quantified as the slope of a linear regression curve of the decay curve. Construct the initial root node dataset , and according to Sort the dataset by size; Construct a standard statistic, take the midpoint of the traffic volume observation value in node D as a candidate split point, and divide each candidate split point into left and right subsets, calculate the information gain at any candidate split point, and select the candidate point with the maximum information gain as the optimal split point; Perform significance test on the left subset and the right subset of the child node respectively, if the p-value of the chi-square test is <0.05, then find the split point again, otherwise stop splitting; Sort the tree structure split points and take the integer to obtain the traffic volume grouping boundary.

[0005] In some embodiments, after constructing the annual average daily traffic volume condition inference tree of the area to be evaluated and obtaining the traffic volume grouping boundary, further comprising: Perform within-group homogeneity test on each group of traffic volume samples to verify that the pavement performance degradation rate within the same traffic volume grouping is highly consistent; Perform between-group heterogeneity test on each group of traffic volume samples to verify that there are significant differences in the degradation rates of different traffic volume groupings.

[0006] In some embodiments, based on the traffic volume grouping boundary, the road network of the area to be evaluated is screened, and target road segments of different traffic volume groupings are selected, specifically: Screen the road network of the area to be evaluated, filter out abnormal values by implementing sliding window filtering processing on gantry data, and select road segments with service time greater than a preset period; Based on the equivalent traffic volume conversion model considering axle load spectrum, the annual average daily traffic volume of the road segment is calculated; According to the traffic volume classification and the specific pavement performance evaluation index, the inspection data of each expressway road segment is grouped, Bootstrap resampling method is used to randomly extract n observation values with replacement, the grouping sample is determined, and the grouping sample includes fitting group and external validation group, and the sample distribution consistency of each grouping is verified.

[0007] In some embodiments, based on the composite pavement performance degradation function, a PPI degradation curve is constructed, and based on the PPI degradation curve, a maintenance time window is obtained, specifically: Preliminarily set the terminal value of the service period performance , after S-curve fitting, based on Monte-Carlo method, the fitting group data is sampled for times, the fitting parameters with the smallest residual sum of squares and the largest determination coefficient are selected, the fitting curve is obtained, the service period of 63.2% degradation state in the fitting curve and the upper limit value of the service period performance are extracted. With and as the limiting conditions, after the exponential curve fitting, the fitting group data is sampled based on the Monte Carlo method Parameter sampling, screening the fitting parameters with the smallest residual sum of squares and the largest coefficient of determination , to obtain the fitting curve , extract the road performance index when the service life is 15 years in the fitting curve ; With as the final value of the road service period performance, repeat the S-shaped curve fitting to obtain the fitting curve ; The fitting curve and the fitting curve are synthesized by using the weighted average method, and the fitting curve and the fitting curve are determined by L-M algorithm regression analysis based on the fitting group data to determine the weight, and the composite road performance degradation equation is obtained. On the basis of the composite attenuation curve fitting, the third derivative of the combined curve corresponds to The starting point of performance accelerated decay, the second derivative corresponds to The curvature extremum point, construct the maintenance time window .

[0008] In some embodiments, the composite road performance degradation equation is specifically: ; Wherein, is the initial value of the performance index; is the final value of the performance index; , is the weight coefficient; , is the S-shaped curve model parameter; is the initial value of the performance index, usually 98-100, indicating the initial road performance when the newly built road is completed and opened to traffic; , is the exponential curve model parameter.

[0009] In some embodiments, based on the target function and the traffic volume classification, the recommended inspection cycle of different traffic volume groups is determined, specifically: A dataset is created by extracting the time from the preventive maintenance threshold point to the preventive maintenance critical value in selected road segments according to traffic volume levels. A likelihood function for the missed detection risk probability in the objective function of the Weibull distribution is constructed, and maximum likelihood estimation is performed. Back-substitution is used to determine the constraint equation for the risk of missed detection, and the objective function form is obtained; According to different time intervals Calculate the corresponding comprehensive testing cost during the operation period, and find the testing interval that minimizes the comprehensive testing cost during the operation period. Will As a suggested initial testing time, the testing interval that minimizes the overall testing cost during the operational period will be used as the recommended scheduled maintenance cycle. This represents the year corresponding to the starting point of accelerated performance degradation.

[0010] Accordingly, the present invention also proposes a low-traffic-volume road classification detection and evaluation device based on pavement performance prediction, the device comprising: The construction module is used to build a conditional inference tree for the annual average daily traffic volume of the region to be evaluated, and to obtain the traffic volume grouping boundaries. The screening module is used to screen the road network of the area to be evaluated based on the traffic volume grouping boundary and select target road segments of different traffic volume groups. The acquisition module is used to construct a PPI decay curve based on the composite pavement performance decay function, and to obtain a maintenance time window based on the PPI decay curve; The constraint module is used to construct an objective function regarding the comprehensive inspection cost during the operation period based on the maintenance time window, and adopts a reliability-considering approach. The Weibull distribution constrains the probability of missed detection in the objective function; The scheduled maintenance module is used to determine the recommended scheduled maintenance cycle for different traffic volume groups based on the objective function and the traffic volume classification.

[0011] One embodiment of the present invention also provides a computing device, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, wherein when the computer-executable instructions are executed by the processor, the steps of the low-traffic-volume road classification detection and evaluation method based on pavement performance prediction as described above are implemented.

[0012] One embodiment of the present invention also provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the low-traffic-volume road classification detection and evaluation method based on pavement performance prediction as described above.

[0013] By applying the above technical scheme, a low-traffic road grading detection evaluation method based on road performance prediction is proposed, which comprises the following steps: constructing an annual average daily traffic condition inference tree of an area to be evaluated to obtain traffic grouping boundaries; performing road network screening on the area to be evaluated based on the traffic grouping boundaries to select target road sections of different traffic groupings; constructing a PPI decay curve based on a composite road performance decay function, and obtaining a maintenance time window based on the PPI decay curve; constructing a target function about comprehensive detection cost during the operation period based on the maintenance time window, and constraining the missed detection risk probability in the target function by using a Weibull distribution considering reliability ; and determining recommended detection periods of different traffic groupings based on the target function and the traffic grading, thereby solving the problem that the grading detection method for low-traffic roads in the prior art cannot form a complete comprehensive evaluation system. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0015] Figure 1 is a flowchart of a low-traffic road grading detection evaluation method based on road performance prediction provided by an embodiment of the present application; Figure 2 is a structural schematic diagram of a low-traffic road grading detection evaluation device based on road performance prediction provided by an embodiment of the present application; Figure 3 is a structural block diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0016] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application. However, the present application can be practiced in many different ways without departing from the spirit and scope of the present application, and those skilled in the art can make similar modifications without departing from the spirit and scope of the present application, so the present application is not limited to the specific implementation disclosed below.

[0017] The terminology used in this disclosure of one or more embodiments is for the purpose of describing particular embodiments only and is not intended to be limiting of one or more embodiments of the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0018] It is to be understood that, although the terms first, second, etc. can be used herein to describe various information, these terms are not intended to denote a particular order or relationship between the information. These terms are used only to distinguish one from another. For example, a first item can be termed a second item, and, similarly, a second item can be termed a first item, without departing from the scope of one or more embodiments of the present disclosure. As used herein, the term "if' can be interpreted to mean "when" or "upon" or "in response to determining" taking into account the context in which the term is used.

[0019] In particular, as shown in Figures 1-3 , the present application proposes a low-traffic road classification detection and evaluation method based on road performance prediction, which comprises the following steps: Step S101, constructing a condition inference tree of the annual average daily traffic volume of the area to be evaluated to obtain the traffic volume grouping boundary.

[0020] In one possible implementation, the condition inference tree of the annual average daily traffic volume of the area to be evaluated is constructed to obtain the traffic volume grouping boundary, specifically: According to the condition inference framework, a decision tree is constructed, and the log-standardized traffic volume is selected as a continuous prediction variable, and the road performance change trend is selected as a response variable, and the change trend is quantified as the linear regression slope of the decay curve , an initial root node data set is constructed , and the data set is sorted according to size; standard statistics are constructed, and the midpoint of the traffic volume observation value in the node D is taken as a candidate split point, and each candidate split point is divided into left and right subsets, the information gain at any candidate split point is calculated, and the candidate point with the maximum information gain is selected as the optimal split point; The left subset and the right subset of the child node are subjected to significance test respectively, if the p-value of the chi-square test is <0.05, the split point is re-found, otherwise the split is stopped; After sorting the tree structure split points, the traffic volume grouping boundary is obtained by rounding.

[0021] In a possible implementation, after constructing the annual average daily traffic condition inference tree of the region to be evaluated and obtaining the traffic volume grouping boundary, the following is further included: Performing within-group homogeneity test on each group of traffic volume samples to verify that the pavement performance degradation rates within the same traffic volume grouping have high consistency; Performing between-group heterogeneity test on each group of traffic volume samples to verify that there are significant differences in the degradation rates of different traffic volume groupings.

[0022] Specifically, the following two points need to be met when dividing the traffic volume: (1) Within the same traffic volume classification, the trend of change of pavement performance indicators (i.e., the degradation curve) over time for different road sections is as consistent as possible; (2) The differences in the degradation curves between different traffic volume classifications are as large as possible.

[0023] Therefore, a conditional inference tree (CTREE) needs to be introduced. The conditional inference tree is a recursive partitioning method based on statistical testing, which can automatically handle continuous predictor variables, select split points through hypothesis testing, and ensure the statistical significance of splitting. Its core advantage is that it can directly test the correlation between the predictor variable and the response variable, and only split when the correlation is significant, making the grouping more reliable. The operation process is as follows: 1. Construct a decision tree according to the conditional inference framework, and select the log-standardized traffic volume as the continuous predictor variable:

[0024] Select the trend of change in pavement performance as the response variable, and for simplicity, quantify the trend of change as the linear regression slope of the degradation curve , construct the initial root node dataset , and sort the dataset according to ; 2. Construct the standard statistic:

[0025] wherein, the statistic

[0026] the sample mean ; the sample variance

[0027] 3. Take the midpoint of the traffic volume observation values in node D (initially the root node) as the candidate split point:

[0028] Each can be divided into left and right subsets: left subset

[0029] right subset

[0030] Compute information gain at any candidate split point:

[0031] where, , , , are linear regression slope of i-th sample, D set, left subset and right subset respectively; , are number of nodes of left and right subset respectively.

[0032] Select the candidate point with the largest as the optimal split point.

[0033] 4, Set up null hypothesis : and are independent. Perform significance test on the left and right subsets of child nodes and respectively. If , reject and repeat step (3) to continue the process of finding split points. Otherwise, stop splitting.

[0034] 5, Sort the tree structure split points and take the integer part to get the traffic volume grouping boundary.

[0035] 6, Perform within-group homogeneity test on each group of traffic volume samples to verify that the road performance decay rates within the same traffic volume grouping have high consistency: Levene's test of homogeneity of variance: (null hypothesis H0: all group variances are equal) Within-group correlation coefficient:

[0036] where, , , are the average deviation of the j-th group data, the total average deviation and the average deviation of each sample within the group respectively; , are the between-group sum of squares and the within-group sum of squares respectively; m is the average number of samples in each group; When Levene's test of homogeneity of variance p > 0.05 and the correlation coefficient ICC > 0.8, the within-group homogeneity test is determined to pass, if p > 0.05 and ICC < 0.8, increase the grouping; 7. Inter-group heterogeneity test is performed on each traffic volume sample to verify that there is a significant difference in the decay rate of different traffic volume groups: One-way ANOVA: (Hypothesis H0: all group means are equal) Effect size:

[0037] When one-way ANOVA p < 0.01 and effect size > 0.25, the inter-group heterogeneity test is determined to pass.

[0038] In step S102, the road network of the to-be-evaluated area is screened based on the traffic volume grouping boundary, and target road segments of different traffic volume groups are selected.

[0039] In one possible implementation, the road network of the to-be-evaluated area is screened based on the traffic volume grouping boundary, and target road segments of different traffic volume groups are selected, specifically as follows: The road network of the to-be-evaluated area is screened, and road segments with a service time greater than a preset time limit are filtered out, and the gantry data is subjected to sliding window filtering processing to remove abnormal values; The annual average daily traffic volume of the road segment is calculated based on an equivalent traffic volume conversion model considering axle load spectrum; According to traffic volume classification and specific pavement performance evaluation indexes, the inspection data of each expressway road segment is grouped, Bootstrap resampling method is used to randomly extract n observation values with replacement, grouping samples are determined, the grouping samples include fitting groups and external validation groups, and sample distribution consistency of each group is verified.

[0040] In this embodiment, the road network of the to-be-evaluated area is screened, and road segments with a service time greater than 5 years are filtered out, and the gantry data is subjected to sliding window filtering processing to remove abnormal values; the annual average daily traffic volume of the road segment is calculated based on an equivalent traffic volume conversion model considering axle load spectrum.

[0041] Specifically, according to traffic volume classification and specific pavement performance evaluation indexes, the inspection data of each expressway road segment is grouped, Bootstrap resampling method (repeated 1000 times) is used to randomly extract n observation values with replacement, grouping samples are determined, that is, fitting groups and external validation groups, and sample distribution consistency (P > 0.1) of each group is verified.

[0042] Step S103: Construct a PPI decay curve based on the composite pavement performance decay function, and obtain a maintenance time window based on the PPI decay curve.

[0043] In one possible implementation, a PPI decay curve is constructed based on a composite pavement performance decay function, and a maintenance time window is obtained based on the PPI decay curve, specifically as follows: Preliminary setting of final performance values ​​during service life After fitting the S-curve, the fitted data were analyzed using the Monte Carlo method. Secondary parameter sampling, selecting the parameters with the smallest sum of squared residuals and the smallest coefficient of determination. By determining the fitting parameters, the fitting curve is obtained. Extract the service life of the 63.2% attenuation state from the fitted curve. and service life performance limit ; by and To constrain the conditions, after fitting the exponential curve, the fitted data were analyzed using the Monte Carlo method. Secondary parameter sampling, selecting the parameters with the smallest sum of squared residuals and the smallest coefficient of determination. By determining the fitting parameters, the fitting curve is obtained. The pavement performance indicators used in the fitted curve for a 15-year lifespan were extracted. ; by As the final performance value during the road's service life, the S-shaped curve fitting was repeated to obtain the fitted curve. ; The weighted average method was used to fit the curve. and fitted curve Synthesize the data and refine the fitted curve based on the fitted group data. and fitted curve The weights were determined by performing regression analysis using the LM algorithm, and the performance degradation equation of the composite pavement was obtained. Based on the fitting of the composite attenuation curve, the combined curve is calculated. The third derivative corresponding The second derivative is the starting point for accelerated performance degradation. corresponding Construct a maintenance window for the curvature extreme point. .

[0044] In this embodiment, m data points are randomly selected from the fitted group to establish an S-shaped pavement performance index (PPI) decay model curve that takes into account traffic volume classification correction.

[0045] PPI decay curves are constructed using the Boltzmann function:

[0046] wherein, is the performance index initial value; is the performance index final value; , is the S-shaped curve model parameter. The performance final value in service period is initially set as Based on the Monte-Carlo method, the fitting group data is sampled for times of parameter sampling, and the fitting curve with the minimum residual sum of squares (RSS) and the largest coefficient of determination is screened after fitting. The service life of the 63.2% decay state in the fitting curve and the upper limit value of the performance in service period are extracted; m pieces of data are randomly extracted from the fitting group, and an exponential pavement performance index (PPI) decay model considering traffic volume grading correction is established.

[0047] An exponential function is used to construct the PPI decay curve:

[0048] wherein, is the performance index initial value, usually 98-100, indicating the initial pavement performance when the newly built road is completed and opened to traffic; , is the exponential curve model parameter. Based on the Monte Carlo method, the fitting group data is sampled for times of parameter sampling, and the fitting curve with the minimum residual sum of squares (RSS) and the largest coefficient of determination is screened after fitting. The pavement performance index when the service life is 15 years in the fitting curve is extracted; Based on the fitting group data, the step (5) S-shaped curve fitting is repeated with as the performance final value in service period, to obtain the fitting curve ; A composite pavement performance decay equation is established.

[0049]

[0050] The weighted average method is used to synthesize and , and the L-M algorithm regression analysis is used to determine the weight based on the fitting group data.

[0051] The core of preventive maintenance is to grasp the maintenance threshold window before the performance accelerated degradation, to maintain the pavement in good condition at a lower cost, and to prolong the service life. The composite pavement performance degradation equation shows that the degradation function is a monotonically decreasing nonlinear function, the curve initially gently declines, then steeply accelerates, and finally again gently declines, gradually tending to the lower limit of the performance index.

[0052] The first derivative of the performance index function can represent the instantaneous decay rate, the second derivative can represent the change rate of the decay rate, which is used to describe whether the performance deterioration is accelerating or decelerating, and the third derivative can represent the trend of the decay acceleration, which is used to describe whether the acceleration or deceleration deterioration is increasing or decreasing.

[0053] When the second derivative first reaches zero, the performance decay rate first reaches a maximum value, and before and after this point, the performance changes from accelerated deterioration to decelerated deterioration. This point corresponds to the upper limit of the preventive maintenance time limit or the lower limit of the performance threshold, respectively. If the maintenance time is later than this limit, the pavement has suffered more serious damage, and the repair cost increases significantly. At the same time, when the performance index is below this threshold, although the decay rate has slowed down, the overall pavement condition is poor and is not suitable for preventive maintenance.

[0054] When the third derivative first reaches zero, the performance index decay first shows a significant acceleration trend, and before and after this point, the trend of pavement performance decay deterioration begins to increase. This point corresponds to the lower limit of the preventive maintenance time limit or the upper limit of the performance threshold, respectively. This point represents the "warning point" that the performance is about to enter the rapid decline period. At this point, the pavement performance is still at a good level, and if preventive maintenance is performed at this time, the maximum benefit of significantly delaying performance decline and prolonging service life can be obtained at the minimum cost.

[0055] Therefore, based on the fitting of the composite decay curve, the third derivative of the combined curve is calculated The corresponding is the performance accelerated decay starting point (preventive maintenance threshold point), and the second derivative corresponds to is the curvature extreme point (maintenance time limit critical point), and the maintenance time window is constructed.

[0056] For general roadbed sections, based on the analysis of pavement diseases and combined with core drilling data, the treatment method of "surface layer disease treatment + overlay" is preferred. With the increase of traffic volume, asphalt concrete surface layer can be added after the pavement to prolong the overall service life of the pavement. For pavements with large uneven settlement, to reduce diseases such as bump and deflection, a flexible base can be added as a leveling layer to restore smooth pavement alignment.

[0057] In one possible implementation, the composite pavement performance degradation equation is specifically: ; wherein, is the initial value of the performance index; is the final value of the performance index; , is the weight coefficient; , is the parameter of the S-shaped curve model; is the initial value of the performance index, usually 98-100, indicating the initial road surface performance when the newly built road is completed and opened to traffic; , is the parameter of the exponential curve model.

[0058] Step S104, constructing a target function about the comprehensive detection cost in the operation period based on the maintenance time window, and constraining the missed detection risk probability in the target function by using the Weibull distribution considering the reliability.

[0059] Step S105, determining the recommended detection cycle of different traffic groups based on the target function and the traffic classification.

[0060] In one possible implementation, the recommended detection cycle of different traffic groups is determined based on the target function and the traffic classification, specifically: The time from the preventive maintenance threshold point to the preventive maintenance critical value in the selected road section is extracted according to the traffic classification to establish a data set. The likelihood function of the missed detection risk probability in the target function of the Weibull distribution is constructed, and maximum likelihood estimation is performed; The back substitution determines the missed detection risk constraint equation, and the target function form is obtained; The corresponding comprehensive detection cost in the operation period is calculated according to different time intervals respectively, and the detection interval that minimizes the comprehensive detection cost in the operation period is solved; The is taken as the recommended first detection time, and the detection interval that minimizes the comprehensive detection cost in the operation period is taken as the recommended detection cycle, wherein, is the time year corresponding to the performance acceleration decay starting point.

[0061] In this embodiment, road performance detection is the premise of preventive maintenance decision, but frequent detection leads to a sharp increase in operation and maintenance cost, so the detection cycle needs to be balanced from the risk caused by missed detection and the economy of maintenance. The model is constructed based on the following assumptions: (1) the road surface performance decay pattern is mainly related to the traffic load level; (2) the road surface performance can be restored to the initial operation level after timely preventive maintenance; (3) the performance decay pattern after maintenance is consistent with the original performance decay pattern. The model is constructed as follows: ​1. Construct the objective function:

[0062]

[0063] wherein, is the comprehensive detection cost during operation period (ten thousand yuan / km); is the design life of pavement (years); is the detection interval (years); is the probability of missed detection risk; is the single detection cost (ten thousand yuan / km); is the missed detection repair cost (ten thousand yuan / km), which is the difference between the single pavement repair cost and the single preventive maintenance cost; Missed detection risk probability is defined as the probability that the performance index decreases to the preventive maintenance critical value (corresponding to the preventive maintenance timeliness critical point ) within the time interval .

[0064] wherein, is the maximum acceptable failure probability; The missed detection risk probability control constraint uses the Weibull distribution considering reliability:

[0065] wherein, is the scale parameter of Weibull distribution, reflecting the characteristic life, which is the time of 63.2% sample failure; is the shape parameter of Weibull distribution, controlling the failure mode.

[0066] 2. According to the traffic volume classification, extract the time from decreasing to to establish a data set in the selected road section. Construct the likelihood function of Weibull distribution and perform maximum likelihood estimation: ​​​​

[0067] Separately and The maximum likelihood function is obtained using Newton's iterative method based on the dataset. The extreme values ​​are obtained by solving for the Weibull distribution parameters. ; 3. Back-substitution to determine the constraint equation for the risk of missed detection And obtain the objective function form:

[0068]

[0069]

[0070] 4. Order in sequence Calculate separately Solving for the solution minimizes the overall testing cost during the operational period. ; 5. As a recommended time for the first test, As a recommended maintenance cycle In summary, this invention proposes a method for classifying and evaluating low-traffic-volume roads based on pavement performance prediction. The method includes: constructing a conditional inference tree for the annual average daily traffic volume of the area to be evaluated, obtaining traffic volume group boundaries; screening the road network of the area to be evaluated based on the traffic volume group boundaries, selecting target road segments for different traffic volume groups; constructing a PPI decay curve based on a composite pavement performance decay function, and obtaining a maintenance window based on the PPI decay curve; constructing an objective function regarding the comprehensive inspection cost during the operation period based on the maintenance window, and employing a reliability-considered approach. The Weibull distribution constrains the probability of missed detection in the objective function; based on the objective function and the traffic volume classification, the recommended inspection cycle for different traffic volume groups is determined, thereby solving the problem that the traditional classification detection method for low traffic volume roads has failed to form a complete comprehensive evaluation system.

[0071] This application also proposes a low-traffic-volume road classification detection and evaluation device based on pavement performance prediction, such as... Figure 2 As shown, the device includes: Module 10 is used to construct the conditional inference tree of the annual average daily traffic volume of the area to be evaluated, and to obtain the traffic volume grouping boundary; Screening module 20 is used to screen the road network of the area to be evaluated based on the traffic volume grouping boundary and select target road segments of different traffic volume groups; The acquisition module 30 is used to construct a PPI decay curve based on the composite pavement performance decay function, and to obtain a maintenance time window based on the PPI decay curve. a constraint module 40 configured to construct a target function regarding comprehensive detection cost in operation period based on the maintenance time window, and to constrain a missed detection risk probability in the target function by considering a Weibull distribution of reliability; a periodic inspection module 50 configured to determine recommended periodic inspection cycles for different traffic volume groups based on the target function and the traffic volume classification.

[0072] Figure 3 A structural block diagram of a computing device 400 according to one embodiment of the present specification is shown. The components of the computing device 400 include, but are not limited to, a memory 410 and a processor 420. The processor 420 is connected to the memory 410 through a bus 430, and a database 450 is used to save data.

[0073] The computing device 400 also includes an access device 440, which enables the computing device 400 to communicate via one or more networks 460. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 440 can include one or more of any type of network interface (e.g., a network interface card (NIC)), wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC).

[0074] In one embodiment of the present specification, the above-mentioned components of the computing device 400 and other components not shown in the above-mentioned components can also be connected to each other, for example, through a bus. It should be understood that Figure 3 the structural block diagram of the computing device shown is only for the purpose of example, and is not a limitation on the scope of the present specification. Other components can be added or replaced as needed by those skilled in the art. Figure 3 ​​

[0075] The computing device 400 can be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 400 can also be a mobile or stationary server.

[0076] The processor 420 is configured to execute computer-executable instructions to implement the steps of the method for detecting and evaluating low-traffic road classification based on road surface performance prediction. The above is a schematic solution of the computing device of the embodiment. It should be noted that the technical solution of the computing device and the technical solution of the method for detecting and evaluating low-traffic road classification based on road surface performance prediction belong to the same concept, and the details of the technical solution of the computing device not described in detail can be referred to the description of the technical solution of the method for detecting and evaluating low-traffic road classification based on road surface performance prediction.

[0077] An embodiment of the present specification also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the method for detecting and evaluating low-traffic road classification based on road surface performance prediction.

[0078] The above is a schematic solution of the computer-readable storage medium of the embodiment. It should be noted that the technical solution of the storage medium and the technical solution of the method for detecting and evaluating low-traffic road classification based on road surface performance prediction belong to the same concept, and the details of the technical solution of the storage medium not described in detail can be referred to the description of the technical solution of the method for detecting and evaluating low-traffic road classification based on road surface performance prediction.

[0079] An embodiment of the present specification also provides a computer program, which, when executed in a computer, causes the computer to perform the steps of the method for detecting and evaluating low-traffic road classification based on road surface performance prediction.

[0080] The above is a schematic solution of the computer program of the embodiment. It should be noted that the technical solution of the computer program and the technical solution of the method for detecting and evaluating low-traffic road classification based on road surface performance prediction belong to the same concept, and the details of the technical solution of the computer program not described in detail can be referred to the description of the technical solution of the method for detecting and evaluating low-traffic road classification based on road surface performance prediction.

[0081] The above-described embodiments of the application have several aspects, no single one of which is solely responsible for the application's desirable attributes. Without limiting the scope of the application as expressed by the claims which follow, some further embodiments make these aspects even more useful. Other embodiments can result in less desirable attributes.

[0082] The computer readable medium can include any entity or apparatus capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier signal, telecommunication signal, software distribution medium, etc. It should be noted that the computer readable medium can include appropriate contents according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.

[0083] It should be noted that for the foregoing method embodiments, the acts described therein can be performed in a different order from that described, and that some acts can be performed in parallel or concurrently. It will also be understood that the above-described method embodiments are only preferred embodiments, and the acts and modules involved are not necessarily all required by the above-described method embodiments.

[0084] In the above embodiments, the description of each embodiment is focused on, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0085] The above disclosed preferred embodiments of the present application are only used to help explain the present application. Alternative embodiments do not describe all the details, nor limit the application to only the specific embodiments described. Obviously, according to the content of the embodiments of the present application, many modifications and changes can be made. The present application selects and specifically describes these embodiments in order to better explain the principles and practical application of the embodiments of the present application, so that those skilled in the art can well understand and use the present application. The present application is limited by the claims and their full scope and equivalents.

Claims

1. A method for classifying and evaluating low-traffic roads based on pavement performance prediction, characterized in that, The method includes: Construct a conditional inference tree for the annual average daily traffic volume of the region to be evaluated to obtain the traffic volume grouping boundaries; Based on the traffic volume grouping boundaries, road network screening is performed on the area to be evaluated, and target road segments with different traffic volume groups are selected. A PPI decay curve is constructed based on a composite pavement performance decay function, and a maintenance time window is obtained based on the PPI decay curve. Based on the maintenance time window, an objective function for the comprehensive inspection cost during the operation period is constructed, and the probability of missed detection in the objective function is constrained by a Weibull distribution that considers the reliability β. Based on the objective function and the traffic volume classification, the recommended scheduled maintenance cycle for different traffic volume groups is determined.

2. The method according to claim 1, characterized in that, Construct a conditional inference tree for the annual average daily traffic volume of the region to be evaluated to obtain the traffic volume grouping boundaries, specifically: A decision tree is constructed based on the conditional inference framework, and the log-standardized traffic volume is selected. As a continuous predictor variable, the trend of pavement performance change is selected as the response variable, and the trend is quantified as the slope of a linear regression curve of the decay curve. Construct the initial root node dataset , and according to Sort the dataset by size; Construct standard statistics, take the midpoint of traffic volume observations within node D as candidate split points, divide each candidate split point into left and right subsets, calculate the information gain at any candidate split point, and select the candidate point with the largest information gain as the optimal split point; child nodes left subset and right subset Perform significance tests separately. If the p-value of the chi-square test is <0.05, then search for split points again; otherwise, stop splitting. The traffic volume grouping boundary is obtained by sorting the split points of the tree structure and rounding them down.

3. The method according to claim 2, characterized in that, After constructing the conditional inference tree for the annual average daily traffic volume of the area to be evaluated and obtaining the traffic volume group boundaries, the following steps are also included: Intra-group homogeneity tests were conducted on traffic volume samples in each group to verify that the pavement performance degradation rate was highly consistent within the same traffic volume group. Intergroup heterogeneity was tested on the traffic volume samples to verify that there were significant differences in the decay rate of different traffic volume groups.

4. The method according to claim 3, characterized in that, Based on the traffic volume grouping boundaries, a road network screening is performed on the area to be evaluated, selecting target road segments for different traffic volume groups, specifically: The road network of the area to be evaluated is screened to identify road sections that have been in service for more than the preset number of years. Sliding window filtering is applied to the gantry data to remove outliers. Based on the equivalent traffic volume conversion model that takes into account the axle load spectrum, the annual average daily traffic volume of this road section is calculated. According to traffic volume classification and specific pavement performance evaluation indicators, the regular inspection data of each highway section are grouped. The Bootstrap resampling method is used to randomly select n observations with replacement to determine the grouped samples. The grouped samples include a fitting group and an external validation group, and the consistency of the sample distribution of each group is verified.

5. The method according to claim 4, characterized in that, A PPI decay curve is constructed based on a composite pavement performance decay function, and a maintenance time window is obtained based on the PPI decay curve, specifically as follows: Preliminary setting of final performance values ​​during service life After fitting the S-curve, the fitted data were analyzed using the Monte Carlo method. Secondary parameter sampling, selecting the parameters with the smallest sum of squared residuals and the smallest coefficient of determination. By determining the fitting parameters, the fitting curve is obtained. Extract the service life of the 63.2% attenuation state from the fitted curve. and service life performance limit ; by and To constrain the conditions, after fitting the exponential curve, the fitted data were analyzed using the Monte Carlo method. Secondary parameter sampling, selecting the parameters with the smallest sum of squared residuals and the smallest coefficient of determination. By determining the fitting parameters, the fitting curve is obtained. The pavement performance indicators used in the fitted curve for a 15-year lifespan were extracted. ; by As the final performance value during the road's service life, the S-shaped curve fitting was repeated to obtain the fitted curve. ; The weighted average method was used to fit the curve. and fitted curve Synthesize the data and refine the fitted curve based on the fitted group data. and fitted curve The weights were determined by performing regression analysis using the LM algorithm, and the performance degradation equation of the composite pavement was obtained. Based on the fitting of the composite attenuation curve, the combined curve is calculated. The third derivative corresponding The second derivative is the starting point for accelerated performance degradation. corresponding Construct a maintenance window for the curvature extreme point. .

6. The method according to claim 5, characterized in that, The composite pavement performance degradation equation is as follows: ； in, These are the initial values ​​for the performance metrics; This refers to the final value of the performance metric. , These are the weighting coefficients; , These are the parameters for the S-curve model; This is the initial value of the performance index, usually taken as 98-100, representing the initial pavement performance of a newly built road when it is completed and opened to traffic. , These are the parameters for the exponential curve model.

7. The method according to claim 1, characterized in that, Based on the objective function and the traffic volume classification, the recommended maintenance cycle for different traffic volume groups is determined as follows: Data sets are established by extracting the time from the preventive maintenance threshold point to the preventive maintenance critical value in the selected road segments according to traffic volume classification. A likelihood function of the missed detection risk probability in the objective function of the Weibull distribution is constructed, and maximum likelihood estimation is performed. Back-substitution determines the constraint equation for the risk of missed detection, and the objective function form is obtained; According to different time intervals Calculate the corresponding comprehensive testing cost during the operation period, and find the testing interval that minimizes the comprehensive testing cost during the operation period. Will As a suggested initial testing time, the testing interval that minimizes the overall testing cost during the operational period will be used as the recommended scheduled maintenance cycle. This represents the year corresponding to the starting point of accelerated performance degradation.

8. A low-traffic road classification detection and evaluation device based on pavement performance prediction, characterized in that, The device includes: The construction module is used to build a conditional inference tree for the annual average daily traffic volume of the region to be evaluated, and to obtain the traffic volume grouping boundaries. The screening module is used to screen the road network of the area to be evaluated based on the traffic volume grouping boundary and select target road segments of different traffic volume groups. The acquisition module is used to construct a PPI decay curve based on the composite pavement performance decay function, and to obtain a maintenance time window based on the PPI decay curve; The constraint module is used to construct an objective function regarding the comprehensive inspection cost during the operation period based on the maintenance time window, and adopts a reliability-considering approach. The Weibull distribution constrains the probability of missed detection in the objective function; The scheduled maintenance module is used to determine the recommended scheduled maintenance cycle for different traffic volume groups based on the objective function and the traffic volume classification.

9. A computing device, characterized in that, include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the low-traffic-volume road classification detection and evaluation method based on pavement performance prediction as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores computer-executable instructions, which, when executed by a processor, implement the steps of the low-traffic-volume road classification detection and evaluation method based on pavement performance prediction as described in any one of claims 1 to 7.

Images