Road condition index prediction method and system based on index attenuation

Abstract

The invention relates to a road condition index prediction method and system based on index attenuation. The method comprises the following steps: firstly, obtaining historical and planned maintenance information, road condition data and road surface types of a target road, calculating annual index attenuation values of various road surface types, extracting sample paragraphs of specified road surface types, calculating historical annual interval attenuation values of the sample paragraphs, and further obtaining a latest annual attenuation value through a weighted average algorithm; dividing a target road into basic road units, determining the latest road surface type of each unit by using a time sequence covering method, and matching a corresponding attenuation value; and finally, subtracting the attenuation value from the latest annual road condition index value to obtain the predicted annual road condition index of each unit. According to the method, data-driven attenuation law analysis is adopted, weighted average and time sequence coverage technologies are combined, the accuracy of road surface technical condition index prediction is improved, and maintenance road section selection, scheme making and resource allocation decision making can be effectively guided.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of highway pavement maintenance engineering, and in particular to a road condition index prediction method and system based on index attenuation. BACKGROUND

[0002] At present, the highway network in China has entered the peak period of maintenance. In the whole life cycle management of pavement maintenance engineering, the evaluation indexes of pavement technical condition, such as pavement damage condition index PCI, pavement riding quality index RQI and pavement rutting depth index RDI, play a crucial role. These indexes are the core basis for highway maintenance units to carry out maintenance engineering design, quality detection and engineering implementation, and as an objective reflection of road condition, maintenance demand and maintenance effectiveness, they deeply affect the planning, execution and evaluation links of pavement maintenance work.

[0003] With the continuous expansion of highway maintenance scale and the increase of maintenance fund investment pressure, highway maintenance units pay more and more attention to the long-term benefits and cost performance of pavement maintenance engineering. Whether the core road condition indexes after maintenance (i.e. "post-maintenance road condition indexes") can be accurately predicted before or during the implementation of the engineering has become a core demand for evaluating the effectiveness of the maintenance scheme, optimizing resource allocation and ensuring the effectiveness of maintenance.

[0004] In the prior art, for the prediction of the core indexes of pavement technical condition, the kilometer average method and the specification recommended model method are mainly used to realize it. The former takes an integral kilometer section as the minimum prediction unit, and extrapolates according to the average historical data or simple trend; the latter uses a specific function model (such as exponential / linear model) recommended by the industry specification to fit and estimate the index decay.

[0005] Although the above two methods can predict and estimate the road condition indexes after maintenance to a certain extent, there are still the following shortcomings. On the one hand, the basic data used by the two estimation methods inevitably have the problems of precision fluctuation and abnormal value; on the other hand, both of them do not exclude the "unnatural" influence of road section data caused by maintenance intervention paragraphs (such as overlaying cover, structural repair, etc.), so that the input data of the prediction model are "polluted". These two shortcomings lead to a large deviation between the prediction results provided by the above methods and the actual road condition indexes after maintenance, and this deviation is difficult to be effectively quantified and controlled, thereby limiting the reliability of the prediction results. SUMMARY

[0006] The purpose of the present application is to provide a road condition index prediction method and system based on index attenuation, which can accurately predict the road condition indexes after the implementation of pavement maintenance engineering, greatly improve the reliability of the prediction results, and effectively guide the pavement maintenance decision.

[0007] In order to achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows:

[0008] In a first aspect, the present application provides a road condition index prediction method based on index attenuation, which comprises the following steps:

[0009] Step 01, obtaining the historical maintenance information of the target road, the planned maintenance information of the target prediction year and the historical road condition index data, and extracting the target pavement type and all pavement paragraph samples corresponding to the target pavement type; based on the historical road condition index data of the target road, calculating the road condition index attenuation value x of each pavement paragraph sample corresponding to the target pavement type in different annual intervals;

[0010] Step 02, calculating the road condition index attenuation average value X of each pavement paragraph sample corresponding to the target pavement type based on the road condition index attenuation average value x; according to the preset step length division weight screening interval, counting the number ratio of the target pavement type corresponding to the pavement paragraph sample falling into each weight screening interval, and determining the attenuation calculation weight corresponding to each pavement paragraph sample according to the quantity ratio information;

[0011] Step 03, calculating the latest annual road condition index attenuation value K corresponding to the target pavement type by using the weighted average algorithm;

[0012] Step 04, repeating steps 02 and 03 to calculate the latest annual road condition index attenuation value K of all target pavement types corresponding to the target road;

[0013] Step 05, dividing the target road into several basic road units, and determining the latest pavement type corresponding to each basic road unit based on the historical maintenance information and the planned maintenance information of the target prediction year using the time sequence covering method;

[0014] Step 06, matching the latest pavement type corresponding to each basic road unit with each target pavement type to determine the latest annual road condition index attenuation value K corresponding to such basic road unit;

[0015] Step 07, extracting each road condition index value of each basic road unit in the latest year from the historical road condition index data of the target road, and subtracting the latest annual road condition index attenuation value K applicable to each basic road unit, to finally obtain the target prediction year road condition index information of each basic road unit.

[0016] As a preferred embodiment of the present application, in step 01, the calculation method of the road condition index attenuation average value X of each pavement paragraph sample corresponding to the target pavement type is as follows:

[0017] Taking the year value of the latest year as the end point and the year value of each starting year recorded in the historical maintenance information as the starting point, a plurality of annual intervals with an annual number increasing one by one are formed;

[0018] Based on the arithmetic mean method, the road condition index attenuation value x of each annual interval corresponding to each road section sample of the target road surface type is calculated;

[0019] For each road section sample, the road condition index attenuation values x of all annual intervals corresponding thereto are calculated by the arithmetic mean method to obtain the road condition index attenuation mean value X of the road section sample.

[0020] As a preferred embodiment of the present application, the attenuation value interval includes an interval range upper limit, an interval range lower limit and an interval step; the interval step is equal to a preset step; and the interval range lower limit is zero.

[0021] As a preferred embodiment of the present application, the division method of the basic road unit is specifically: based on three division standards of hundred-meter stake number, road width and lane, the target road is divided into a plurality of basic road units.

[0022] As a preferred embodiment of the present application, the road condition index prediction method based on index attenuation further comprises the following steps of obtaining the target prediction year road condition index information of the kilometer section of the target road:

[0023] The target prediction year road condition index information of all basic road units contained in the to-be-calculated kilometer section is extracted;

[0024] The target prediction year road condition index information of the to-be-calculated kilometer section is calculated and obtained based on the index calculation formula.

[0025] As a preferred embodiment of the present application, the historical maintenance information of the basic road unit includes a historical maintenance measure type and an implementation year.

[0026] As a preferred embodiment of the present application, the road condition index includes a pavement damage condition index PCI, a pavement ride quality index RQI and a pavement rutting depth index RDI.

[0027] As a preferred embodiment of the present application, if the basic road unit has a maintenance plan in the target prediction year, the finally obtained target prediction year road condition index information is the index value that can be reached by the basic road unit after the completion of the maintenance plan.

[0028] In a second aspect, the present application provides a road condition index prediction system based on index attenuation, which comprises:

[0029] a data acquisition and attenuation calculation module, which acquires historical maintenance information of a target road, planned maintenance information of a target prediction year and historical road condition index data, extracts a target pavement type and all pavement section samples corresponding to the target pavement type, and calculates road condition index attenuation values of each pavement section sample corresponding to the target pavement type in different annual intervals based on the historical road condition index data of the target road;

[0030] a weight determination module, which is configured to calculate road condition index attenuation average values of each pavement section sample corresponding to the target pavement type based on the road condition index attenuation average values, divide a weight screening interval according to a preset step length, count a number ratio of the pavement section samples corresponding to the target pavement type falling into each weight screening interval, and determine attenuation calculation weights corresponding to each pavement section sample according to the number ratio information;

[0031] a weighted average calculation module, which is configured to calculate a latest annual road condition index attenuation value corresponding to the target pavement type by using a weighted average algorithm;

[0032] a repeated calculation control module, which is configured to repeatedly call the weight determination module and the weighted average calculation module to calculate latest annual road condition index attenuation values of all target pavement types corresponding to the target road;

[0033] a road unit division and type determination module, which is configured to divide the target road into a plurality of basic road units, and determine latest pavement types corresponding to each basic road unit based on the historical maintenance information and the planned maintenance information of the target prediction year by using a time sequence covering method;

[0034] a matching module, which is configured to match the latest pavement types corresponding to each basic road unit with each target pavement type, and determine latest annual road condition index attenuation values corresponding to such basic road units;

[0035] a prediction calculation module, which is configured to extract each road condition index value of each basic road unit in the latest year from the historical road condition index data of the target road, subtract the latest annual road condition index attenuation value corresponding to each basic road unit, and finally obtain target prediction year road condition index information of each basic road unit.

[0036] In a third aspect, the present application provides an electronic device, which comprises a processor and a memory; the processor is connected with the memory; the memory is used for storing executable program codes; the processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to execute the steps of the above-mentioned road condition index prediction method based on index attenuation.

[0037] In summary, the present application has the following advantages:

[0038] 1. The present application effectively reduces the prediction deviation caused by data precision fluctuation and abnormal values by replacing the traditional model fitting or kilometer average method with a statistical method based on a large amount of actual detection data. The core road condition index decay value and the core road condition index value of the present method are derived from objective historical data statistics rather than theoretical model assumptions, making the prediction results closer to the engineering practice and significantly improving the accuracy and reliability of the prediction.

[0039] 2. The present application innovatively uses the time sequence coverage method to determine the "latest pavement type" of the basic road unit. This key step can effectively identify and exclude the "pollution" of historical maintenance measures (such as overlaying the cover) on the natural decay law of the pavement, ensuring that the data samples used to calculate the index decay can better reflect the natural evolution law of a specific pavement type without major intervention.

[0040] 3. The method provided by the present application does not depend on a specific road condition decay model, the core logic is the road condition index decay law, and the core algorithm is based on data statistics, which can be applied to the prediction of different pavement types and various road condition indexes (such as PCI, RQI, RDI) of highways of all grades.

[0041] 4. The present method can convert complex prediction problems into standardized data processing steps that can be easily implemented and promoted in actual maintenance management systems, providing unified, efficient, and accurate data support for key decision-making links such as maintenance road section selection, scheme development, and fund allocation, and has high practical value and wide applicability. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0043] Figure 1 The flowchart of the present road condition index prediction method;

[0044] Figure 2 The structural block diagram of the present road condition index prediction system. DETAILED DESCRIPTION

[0045] The subject matter described herein will now be discussed with reference to example implementations. It should be understood that these implementations are discussed solely for the purpose of illustrating aspects of the subject matter described herein and are not a limitation of the scope, applicability, or examples set forth in the claims. Changes in the function and arrangement of elements discussed can be made without departing from the scope of the subject matter described herein. Various examples can omit, substitute, or add various procedures or components as appropriate. For instance, the methods described can be performed in an order different than that described, and / or various steps can be added, omitted, or combined. Also, features described with respect to some examples can be combined in other examples.

[0046] As used herein, the term "includes" and its variants are meant to be interpreted broadly. The term "based on" means "based, at least in part, on." The terms "one embodiment" and "an embodiment" mean "at least one embodiment." The term "another embodiment" means "at least one additional embodiment." The terms "a first," "a second," etc. do not require that there be only one of each, but rather, there can be one or more of each. The following detailed description is presented in terms of examples. Further

[0047] As Figure 1 shown, a road condition index prediction method based on index attenuation according to an embodiment of the present application is described in detail below.

[0048] First, the present method can be applied to road condition index prediction of roads of various grades, and the road can include roadbed, bridge, tunnel, and other attribute road sections. The following embodiment is exemplarily shown by taking roadbed attribute as an example. Taking a certain expressway "LQ Expressway" as an example, the present method is used to predict the road condition index of the roadbed of the LQ Expressway at the end of 2025 (the prediction work time is the first half of 2025, and at this time, the index detection work at the end of 2025 has not yet started, i.e., when the prediction is made, there is no road condition index data information in 2025).

[0049] Step 01, obtain historical maintenance information of the target road, planned maintenance information of the target prediction year, and historical road condition index data, and extract the target pavement type and all pavement section samples corresponding to the target pavement type; based on the historical road condition index data of the target road, calculate the road condition index attenuation value x of each pavement section sample corresponding to the target pavement type in different annual intervals;

[0050] Firstly, data acquisition is carried out, historical maintenance information (2020-2024 maintenance measures such as non-surfacing disease treatment, SMA13 surfacing, etc.) of LQ highway, planned maintenance information of target prediction year (2025), historical road condition index data (PCI, RQI, RDI index data in 2020-2024) are collected, the maintenance history, historical data and road section attribute are combined, all pavement types (such as "subgrade-super-thin 20", "subgrade-SMA13 surfacing 20" and the like) are obtained, and the data is cleaned to eliminate abnormal values.

[0051] It should be explained that in the embodiment, the maintenance measure name adopts a simple note name, and the maintenance year takes the last two digits of the year number, such as SMA13 surfacing in 2020, which is simply noted as "SMA13 surfacing 20", and non-surfacing disease treatment planned to be implemented in 2025 is simply noted as "non-surfacing 25", and other measures are marked in a similar manner.

[0052] Then, from all pavement types corresponding to LQ highway, the target pavement type is selected, and all pavement paragraph samples corresponding to the target pavement type are extracted; the length of each pavement paragraph sample is indefinite, which is marked and divided according to the record in the actual database; then, based on the historical road condition index data of the target road, the road condition index decay value x of each pavement paragraph sample corresponding to the target pavement type in different annual intervals is calculated.

[0053] In the application, the road condition index refers to the pavement technical condition index, and the pavement technical condition index provided in the embodiment includes the pavement damage condition index PCI, the pavement driving quality index RQI and the pavement rutting depth index RDI, which will have a clear detection and evaluation value recorded in the road maintenance database at the end of each year for subsequent prediction.

[0054] Step 02, the road condition index decay average value X of each pavement paragraph sample corresponding to the target pavement type is calculated based on the road condition index decay average value x; the number ratio of the target pavement type corresponding to the pavement paragraph sample falling into each weight screening interval is calculated according to the preset step length division weight screening interval, and the decay calculation weight corresponding to each pavement paragraph sample is determined according to the quantity ratio information;

[0055] The year value of the latest year (i.e. the latest year with detection data, which is specifically the year before the target prediction year 2025, i.e. 2024) is taken as the end point, and the year value of the starting year (2020) recorded in the historical maintenance information is taken as the starting point, to form a plurality of annual intervals (i.e. 2020-2024, 2021-2024, 2022-2024, 2023-2024) with annual number increasing one by one.

[0056] The road condition index decay value x corresponding to each annual interval of each road section sample of the target road type is calculated based on the arithmetic mean method; taking the pavement damage condition index PCI and the road type "1 lane, sub-thin 20 on roadbed" as an example, there are a total of 8 road section samples of this road type on the LQ highway that have experienced at least one "1 lane, sub-thin 20 on roadbed" event in the period from 2020 to 2024, and the PCI index of each road section sample in the period from 2020 to 2024 is statistically summarized as shown in Table 1:

[0057] Table 1 .

[0058] Next, the road condition index decay average of the "1 lane, sub-thin 20 on roadbed" road type in the periods from 2020 to 2024, from 2021 to 2024, from 2022 to 2024, and from 2023 to 2024 is calculated.

[0059] The calculation method is to calculate the difference between the PCI in 2020 and the PCI in 2024, the difference between the PCI in 2021 and the PCI in 2024, the difference between the PCI in 2022 and the PCI in 2024, and the difference between the PCI in 2023 and the PCI in 2024 for each road section sample, divide by the number of years, and obtain the average annual decay value, and then calculate the arithmetic mean of the four average annual decay values (and eliminate non-positive values) to obtain a road condition index decay average value X corresponding to each road section sample, which can be denoted as X i (PCI), where i is the code of the road section sample. As shown in Table 2:

[0060] Table 2 .

[0061] Next, the weight selection interval is divided according to the preset step size, the number ratio of the road section samples corresponding to the target road type falling within each weight selection interval is calculated, and the decay calculation weight corresponding to each road section sample is determined according to the number ratio information;

[0062] After obtaining each sample section X i (PCI) in Table 2, the weight selection interval is divided according to the preset step size, in this embodiment, the upper limit of the interval range is 3, the lower limit of the interval range is 0, and the interval step size is 0.2 (the interval step size of RDI and RQI is relatively more precise, and is set to 0.1), then the number of road section samples in each weight selection interval is counted, denoted as n; the number ratio of road section samples in each weight selection interval is calculated, denoted as c; it should be noted that the values less than or equal to 0 or greater than 3 in the PCI decay average are considered as invalid values and do not participate in the statistics, and the upper threshold of the invalid value can be differentiated according to the actual situation of different road sections and different road type indexes. The results obtained in this step are shown in Table 3:

[0063] Table 3 .

[0064] Step 03, using the weighted average algorithm to calculate the latest annual road condition index attenuation value K corresponding to the target road surface type;

[0065] According to the proportion c of the number of road section samples in each weight screening interval, the weight ω corresponding to each weight screening interval is calculated, and the calculation method is to use the c corresponding to the target calculation interval i Divide the sum of each paragraph sample c i , get ω i , where i is the code of the road section sample corresponding to the weight screening interval.

[0066] Next, based on the weight ω i corresponding to the weight screening interval and the PCI attenuation mean X i (PCI) of the corresponding road section sample i, the latest annual road condition index attenuation value K(PCI) is measured and calculated using the weighted average method, where the latest annual road condition index attenuation value K(PCI) represents the attenuation value corresponding to the PCI index in the span of 2024-2025; Specifically, the K value calculation method is to calculate the sum of the product of X i and ω i weighted average, that is, to calculate the sum of the product of X i and ω i , this attenuation value K(PCI) will be used for subsequent prediction calculation, as shown in the following table 4:

[0067] Table 4 .

[0068] Step 04, repeat steps 02 and 03 to calculate the latest annual road condition index attenuation value K (including PCI, RDI and RQI) of all target road surface types corresponding to the target road, as shown in Table 5.

[0069] Table 5 .

[0070] Step 05, divide the target road into several basic road units, and based on the historical maintenance information and the planned maintenance information of the target prediction year, use the time sequence coverage method to determine the latest road surface type corresponding to each basic road unit;

[0071] After obtaining the latest annual PCI attenuation value K corresponding to each road surface type, the final prediction step is performed, in which the target road needs to be divided into several basic road units. Specifically, the target road is divided into several basic road units based on the three division standards of hundred-meter stake number, road width, and lane, and the unit road segment attributes are marked (including three types of roadbed, bridge, and tunnel).

[0072] In this embodiment, only the roadbed attribute is used as an example to illustrate the method steps. In other possible embodiments, the same step method can also be used to predict the index of the bridge, tunnel, or a mixture of any two or all three of them.

[0073] In this embodiment, the LQ highway basic road units are arranged in increasing order of stake number, as shown in Table 6 (only some basic road unit examples are shown):

[0074] Table 6 .

[0075] According to the historical maintenance information, the LQ highway pavement maintenance history started in 2020. The types of pavement maintenance measures implemented include non-coating disease treatment (abbreviated as "non-coating"), SMA13 coating, SMA10 coating, ultra-thin wearing layer (abbreviated as "ultra-thin"), etc. Based on this, the unit maintenance history is marked, including the type and implementation year of the maintenance measures.

[0076] Based on the planned maintenance information, the pavement maintenance plan for the target prediction year (2025) is classified and marked, including the type and implementation year of the maintenance measures. In this embodiment, the LQ highway pavement maintenance plan for 2025 includes two types of non-coating disease treatment (abbreviated as "non-coating") and SMA13 coating.

[0077] Each hundred-meter basic road unit is classified and counted in the table from left to right according to the maintenance history implemented from 2020 to 2024 in the right side of the unit information (in this embodiment, since each basic road unit does not have maintenance history from 2020 to 2022, in order to simplify, the maintenance history from 2020 to 2022 is counted in one column). The 2025 maintenance plan is listed in the rightmost column of the maintenance history. The statistical form is shown in Table 7 (some unit examples are shown, and the "attribute" column in the table refers to the previous example table, which is abbreviated as "unit code" to reduce the table size).

[0078] Table 7 .

[0079] After obtaining the contents in the above table, the maintenance history of each basic road unit 2020-2025 and the maintenance plan in 2025 are logically judged according to time (the judgment principle is that the maintenance measures implemented in the newer year will cover or replace the measures implemented in the earlier year), and then the attributes of each basic road unit, the maintenance history (or maintenance plan), the historical implementation year (or planned implementation year) are combined to obtain the latest pavement type that each 100-meter unit will form after the implementation of the maintenance plan in 2025, as shown in the rightmost column of the following table. Select some basic road units as examples, as shown in the following table (the left column in the "attribute" column refers to the previous example table, and here it is abbreviated as "unit code").

[0080] In Table 8 below, for unit code 1, because there is an implementation plan "non-cover 25" in 2025, the latest pavement type that will be formed after the implementation of the maintenance plan in 2025 is "subgrade-non-cover 25"; for unit code 8, because the unit has not implemented maintenance since 2020 and has no maintenance plan in 2025, therefore, the latest pavement type that will be formed after the implementation of the maintenance plan in 2025 is "subgrade-no maintenance".

[0081] Table 8 .

[0082] Step 06, match the latest pavement type corresponding to each basic road unit with each target pavement type to determine the latest annual road condition index attenuation value K corresponding to such basic road unit;

[0083] Because different target pavement types correspond to different latest annual road condition index attenuation values K, the latest pavement type corresponding to each basic road unit is matched with each target pavement type to determine the latest annual road condition index attenuation value K applicable to each basic road unit, which will be involved in the calculation of Step 07.

[0084] Step 07, extract each road condition index value of each basic road unit in the latest year from the historical road condition index data of the target road, and subtract the latest annual road condition index attenuation value K applicable to each basic road unit, to finally obtain the target predicted annual road condition index information of each basic road unit.

[0085] The purpose of this step is to obtain the predicted values of PCI, RDI and RQI of each basic road unit in the target prediction year (2025) in the target road. First, the types of maintenance measures planned to be implemented in LQ highway in 2025 are determined, including SMA13 overlay and non-overlay disease, and the maintenance measures are combined with the planned implementation road segment attributes (the same as above, and in this embodiment, the subgrade is taken as an example) and the planned implementation lane to obtain the pavement type that will be formed after the implementation of the maintenance plan in LQ highway in 2025, as shown in Table 9 below.

[0086] Table 9 .

[0087] For the obtained road surface type sample section, according to the database information, the detection indicators at the end of 2024 are called, and the PCI, RDI and RQI indicators and their average values of each type of road surface type sample section after the implementation of maintenance measures in 2024 are counted. According to the statistics, for the same maintenance measures, the PCI of 1-lane and 2-lane is almost equal, and the RDI and RQI have slight differences. The statistical results are shown in the following table 10:

[0088] Table 10 .

[0089] Based on table 10 and table 9, the corresponding road surface type of each basic road unit can be determined, and the values of each road condition indicator (PCI, RDI and RQI) of the road surface type in the latest year (2024) are determined. Combined with the latest year road condition indicator decay value K of all road surface types in table 5, the difference calculation can obtain the final target prediction year road condition indicator information of each basic road unit.

[0090] In another possible embodiment, if the basic road unit has a maintenance plan in the target prediction year, the final target prediction year road condition indicator information obtained is the indicator value that the basic road unit can reach after the completion of the maintenance plan.

[0091] The difference from the previous embodiment is that there is a maintenance plan in 2025 and the impact of the maintenance plan on the indicators should be included in the final prediction result. Therefore, when evaluating the final values of each road condition indicator of each basic road unit in 2025, if the basic road unit has a maintenance plan in the target prediction year, the final target prediction year road condition indicator information obtained is the indicator value that the basic road unit can reach after the completion of the maintenance plan. This indicator value more directly reflects the scientific nature of the actual indicators, and in the actual application link, the practicality and applicability of the technical solution provided by this embodiment are stronger.

[0092] In another possible embodiment, the road condition indicator prediction method based on indicator decay further comprises a target road kilometer section target prediction year road condition indicator information acquisition step:

[0093] Extracting the target prediction year road condition indicator information of all basic road units contained in the to-be-calculated kilometer section;

[0094] Obtaining the target prediction year road condition indicator information of the to-be-calculated kilometer section based on the average value calculation method.

[0095] After the target predicted annual road condition index information of each basic road unit is calculated, if the predicted result of a larger range (such as a kilometer section) is needed, the above steps are executed.

[0096] First, from the calculated results, the target predicted annual road condition index information of all basic road units contained in the kilometer section to be calculated is extracted. These information includes the PCI, RQI, RDI predicted values of each hundred-meter unit, and the corresponding DR (pavement damage rate), RD (rut depth), and IRI (international roughness index) values calculated according to the “Highway Technical Condition Evaluation Standard” (JTG 5210-2018).

[0097] Subsequently, the average value calculation method is used for synthesis calculation. Specifically:

[0098] The average value of DR of all (usually 10) basic road units in the kilometer section is calculated as the comprehensive DR value of the kilometer section. Similarly, the average values of RD and IRI of all basic road units in the kilometer section are calculated as the comprehensive RD value and the comprehensive IRI value of the kilometer section.

[0099] Finally, referring to the PCI, RDI, and RQI index calculation formulas provided in the “Highway Technical Condition Evaluation Standard” (JTG 5210-2018), the calculated comprehensive DR value, comprehensive RD value, and comprehensive IRI value of the kilometer section are used as input parameters to calculate the target predicted annual PCI value, RDI value, and RQI value of the kilometer section.

[0100] According to the index calculation principle, the kilometer section PCI, RDI, and RQI index values are calculated from the kilometer section DR, RD, and IRI according to the formula, and the kilometer section DR, RD, and IRI are the DR, RD, and IRI average values of the 10 hundred-meter units corresponding to each kilometer section. Therefore, the DR, RD, and IRI of each hundred-meter unit obtained by the foregoing step-by-step calculation are used to calculate the DR, RD, and IRI average values of each kilometer section, and then the PCI, RDI, and RQI index values of each kilometer section are calculated. The step-by-step calculation principle refers to the corresponding formulas in the “Highway Technical Condition Evaluation Standard” (JTG 5210-2018).

[0101] In another possible embodiment, a road condition index prediction system based on index attenuation is also provided, which includes:

[0102] The data acquisition and attenuation calculation module is configured to acquire historical maintenance information of a target road, planned maintenance information of a target prediction year, historical road condition index data, and all pavement types; select a target pavement type, extract all pavement section samples corresponding to the target pavement type; and calculate road condition index attenuation values x of each pavement section sample corresponding to the target pavement type in different year intervals based on the historical road condition index data of the target road.

[0103] The weight determination module is configured to calculate road condition index attenuation mean values X of each pavement section sample corresponding to the target pavement type according to the road condition index attenuation mean values x, divide a weight screening interval according to a preset step length, count a proportion of the number of the pavement section samples corresponding to the target pavement type falling into each weight screening interval, and determine attenuation calculation weights corresponding to each pavement section sample according to the proportion of the number.

[0104] The weighted average calculation module is configured to calculate a latest year road condition index attenuation value K corresponding to the target pavement type by using a weighted average algorithm.

[0105] The repeated calculation control module is configured to repeatedly call the weight determination module and the weighted average calculation module to calculate the latest year road condition index attenuation values K of all target pavement types corresponding to the target road.

[0106] The road unit division and type determination module is configured to divide the target road into a plurality of basic road units, determine the latest pavement types corresponding to each basic road unit based on the historical maintenance information and the planned maintenance information of the target prediction year by using a time sequence covering method.

[0107] The matching module is configured to match the latest pavement types corresponding to each basic road unit with each target pavement type, and determine the latest year road condition index attenuation values K applicable to each basic road unit.

[0108] The prediction calculation module is configured to extract each road condition index value of each basic road unit in the latest year from the historical road condition index data of the target road, subtract the latest year road condition index attenuation value K applicable to each basic road unit, and finally obtain target prediction year road condition index information of each basic road unit.

[0109] In another possible embodiment, an electronic device is also provided, which includes a processor and a memory; the processor is connected with the memory; the memory is configured to store executable program codes; the processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to execute the steps of the road condition index prediction method based on index attenuation in one of the above embodiments.

[0110] The above described embodiments of the present description have been described. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order and still achieve desirable results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In certain implementations, multitasking and parallel processing can be advantageous.

Claims

1. A road condition index prediction method based on index decay, characterized in that, The methods include: Step 01: Obtain historical maintenance information, planned maintenance information for the target forecast year, and historical road condition index data for the target road, and extract the target pavement type and all pavement segment samples corresponding to the target pavement type; based on the historical road condition index data of the target road, calculate the road condition index decay value of each pavement segment sample corresponding to the target pavement type in different annual intervals. Step 02: Calculate the average attenuation of road condition indicators for each road segment sample corresponding to the target road surface type based on the average attenuation of road condition indicators. Divide the weighted screening interval according to the preset step size, count the proportion of road segment samples corresponding to the target road surface type that fall into each weighted screening interval, and determine the attenuation calculation weight corresponding to each road segment sample based on the proportion information. Step 03: Calculate the latest annual road condition index decay value corresponding to the target road surface type using a weighted average algorithm; Step 04: Repeat steps 02 and 03 to calculate the latest annual road condition index attenuation values ​​for all target road surface types corresponding to the target road. Step 05: Divide the target road into several basic road units. Based on the historical maintenance information and the planned maintenance information for the target predicted year, use the time-series coverage method to determine the latest pavement type corresponding to each basic road unit. Step 06: Match the latest road surface type corresponding to each basic road unit with each of the target road surface types to determine the latest annual road condition index attenuation value corresponding to this type of basic road unit; Step 07: Extract the traffic condition index values ​​for each basic road unit in the latest year from the historical traffic condition index data of the target road, and subtract the applicable traffic condition index attenuation value for the latest year for each basic road unit to obtain the target predicted annual traffic condition index information for each basic road unit. According to claim 1, the method for predicting road condition indicators based on indicator decay is characterized in that, in step 01, the calculation method for the average road condition indicator decay of each road segment sample corresponding to the target road surface type is specifically as follows: Using the latest year as the endpoint and the starting year as the starting year recorded in the historical maintenance information, multiple year intervals are formed with the number of years increasing one by one. The road condition index decay value for each annual interval of each road segment sample corresponding to the target road surface type is calculated based on the arithmetic mean method. For each road segment sample, the arithmetic mean of the road condition index decay values ​​of all corresponding annual intervals is calculated to obtain the mean road condition index decay value of that road segment sample.

2. The road condition index prediction method based on index decay according to claim 2, characterized in that, The attenuation value range includes an upper limit, a lower limit, and a step size; the step size is equal to a preset step size; and the lower limit is zero.

3. The road condition index prediction method based on index decay according to claim 3, characterized in that, The specific method for dividing basic road units is as follows: the target road is divided into several basic road units based on three criteria: 100-meter marker, road width, and lanes.

4. The road condition index prediction method based on index decay according to claim 4, characterized in that, This road condition index prediction method based on index decay also includes the step of obtaining the target road's kilometer-segment target annual road condition index information: Extract the target predicted annual road condition index information of all basic road units contained in the kilometer segment to be calculated; Based on the calculation formulas for each indicator, the target predicted annual road condition indicator information for the kilometer segment to be calculated is obtained.

5. The road condition index prediction method based on index decay according to claim 5, characterized in that, The historical maintenance information of the basic road unit includes the type of historical maintenance measures and the year of implementation.

6. The road condition index prediction method based on index decay according to claim 1, characterized in that, The road condition indicators include the Road Damage Index (PCI), the Road Quality Index (RQI), and the Road Rutting Depth Index (RDI).

7. The road condition index prediction method based on index decay according to claim 1, characterized in that, In S07, if the basic road unit has a maintenance plan for the target forecast year, the final target forecast year road condition index information is the index value that the basic road unit can achieve after the maintenance plan is completed.

8. A road condition index prediction system based on index decay, characterized in that, The system includes: The data acquisition and attenuation calculation module acquires historical maintenance information, planned maintenance information for the target prediction year, and historical road condition index data of the target road, and extracts the target pavement type and all pavement segment samples corresponding to the target pavement type; based on the historical road condition index data of the target road, it calculates the road condition index attenuation value of each pavement segment sample corresponding to the target pavement type in different annual intervals. The weight determination module is used to calculate the average attenuation of road condition indicators for each road segment sample corresponding to the target road surface type based on the average attenuation of road condition indicators; divide the weight screening interval according to the preset step size, count the proportion of road segment samples corresponding to the target road surface type that fall into each weight screening interval, and determine the attenuation calculation weight corresponding to each road segment sample based on the proportion information. The weighted average calculation module is used to calculate the latest annual road condition index decay value corresponding to the target road surface type using a weighted average algorithm. The repeated calculation control module is used to repeatedly call the weight determination module and the weighted average calculation module to calculate the latest annual road condition index decay value for all target road surface types corresponding to the target road. The road unit division and type determination module is used to divide the target road into several basic road units, and based on the historical maintenance information and the planned maintenance information for the target prediction year, use the time-series coverage method to determine the latest pavement type corresponding to each basic road unit. The matching module is used to match the latest road surface type corresponding to each basic road unit with each of the target road surface types to determine the latest annual road condition index decay value corresponding to such basic road units. The prediction calculation module is used to extract the traffic condition index values ​​of each basic road unit in the latest year from the historical traffic condition index data of the target road, and subtract the corresponding traffic condition index decay value of the latest year for each basic road unit to finally obtain the target predicted annual traffic condition index information of each basic road unit.

9. An electronic device, comprising a processor and a memory; the processor being connected to the memory; the memory being used to store executable program code; characterized in that, The processor reads the executable program code stored in the memory to run the program corresponding to the executable program code, so as to perform the steps of the road condition index prediction method based on index decay as described in claim 1.

Images