A water-based road-use polymer road structure design algorithm model

Abstract

This invention discloses a design algorithm model for water-based polymer road structures in the field of engineering technology, including the following core modules: Application scenario-based design module: providing customized structural design schemes for six typical application scenarios, including highway interchanges, heavy-duty mining area roads, secondary highways, urban arterial roads, expressway mainlines, and rural roads; Engineering case database: collecting and managing typical domestic and international engineering case data, including six major typical engineering cases such as the Qingping Interchange of the Tenglong Expressway in Yunnan Province (used for 5 years), the Bozhou section of the Sixu Expressway in Anhui Province (used for 11 years), the Yuba Road in Panzhihua, Sichuan Province (used for 8 years), the X145 Road in Fengdu, Chongqing (used for 7 years), and the S211 Provincial Highway in Ya'an, Sichuan Province (used for 13 years). Through customized design for six typical application scenarios, the accuracy of polymer layer thickness recommendation is improved by more than 30%, effectively avoiding the problems of over-design or under-design.

Description

Technical Field

[0001] This invention relates to the field of engineering technology, specifically to a design algorithm model for polymer road structures used in water-based roads. Background Technology

[0002] Traditional highway pavement base courses mainly use cement-stabilized crushed stone base courses (referred to as "water-stabilized base courses"). Cement-stabilized crushed stone base courses have the following technical problems during long-term use: (1) Reflective cracking problem: Due to the thermal shrinkage and drying shrinkage characteristics of cement hydration products, the base layer is prone to shrinkage cracks during use. These cracks will gradually reflect to the asphalt surface layer, resulting in reflective cracks on the road surface, which seriously affects the service life of the road surface and driving comfort. (2) Water damage problem: Traditional water-stabilized base courses have high permeability, and rainwater can easily seep into the base course, leading to scouring and pumping diseases, which significantly reduces the strength of the pavement structure; (3) High carbon emissions: Cement production is a typical high-carbon emission process. According to calculations by Ergomax (South Africa), an independent third-party assessment agency, the carbon emissions of traditional cement-stabilized crushed stone base courses are approximately 14.23 kg CO2 / m³. 2 The carbon emissions of polymer-stabilized crushed stone base course are only 1.89 kg CO2 / m³. 2 The ratio is approximately 1:7.5; (4) Insufficient economic efficiency: Water-stabilized base courses require a thicker structural layer design, with a total thickness usually above 54cm, while polymer-stabilized crushed stone base courses can control the total thickness to 40-48cm through optimized design, significantly saving material and construction costs.

[0003] Water-based polymer-stabilized crushed stone base course technology is a new type of road base course technology that has been developed in recent years. This technology uses materials such as SRX (polyethyl methacrylate water-based polymer) to stabilize crushed stone, forming a flexible base course with excellent road performance. Compared to traditional cement-stabilized crushed stone base courses, water-based polymer-stabilized crushed stone base courses have significant advantages such as better crack resistance, excellent water stability, high durability, convenient construction, and low carbon footprint.

[0004] However, a complete and systematic design algorithm model for polymer road structures for water-based roads is currently lacking. Existing design methods mainly rely on empirical formulas and analogies, lacking comprehensive optimization analysis of factors such as polymer dosage, structural layer thickness, and mechanical performance indicators, making it difficult to meet the refined design requirements of different application scenarios. Summary of the Invention

[0005] The purpose of this invention is to provide a design algorithm model for polymer road structures for water-based roads, in order to solve the problem of the lack of systematic and scenario-based design methods in the existing technology, and to realize the refined and intelligent design of polymer-stabilized crushed stone base courses for water-based roads.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a design algorithm model for polymer road structures used in water-based roads. The technical solution provided by the present invention includes the following core modules: Application-specific design module: Provides customized structural design solutions for six typical application scenarios, including highway interchanges, heavy-duty mining area roads, secondary highways, urban arterial roads, expressway main lines, and rural roads. Engineering Case Database: Collects and manages data on typical engineering cases from home and abroad, including six major typical engineering cases such as Qingping Interchange of Tenglong Expressway in Yunnan Province (5 years of use), Bozhou Section of Sixu Expressway in Anhui Province (11 years of use), Yuba Road in Panzhihua, Sichuan Province (8 years of use), X145 Road in Fengdu, Chongqing (7 years of use), and S211 Provincial Highway in Ya'an, Sichuan Province (13 years of use). Precise Dosage Gradient Design Module: Based on factors such as climate zone, traffic level, and aggregate type, it provides precise recommendations for polymer dosage and sensitivity analysis; Life cycle cost analysis module: comprehensively considers construction costs, maintenance costs, major and minor repair costs and carbon tax costs, and provides a life cycle economic analysis; Carbon emission refined management module: Based on Ergomax third-party assessment data and S40 Chizhou Expressway case data, it provides carbon emission comparative analysis; The multi-layer elastic system mechanical analysis module automatically calculates six core verification indicators, including road surface acceptance deflection, permanent deformation of asphalt layer, fatigue cracking of asphalt layer, fatigue cracking of polymer base course, penetration strength, and roadbed compressive strain.

[0007] As a further aspect of the present invention: the metrological gradient precise design module includes the following calculation steps: S100: Based on the JTG D50-2017 specification, a five-layer standard water-based road polymer road structure system is constructed, consisting of a fine-grained asphalt concrete surface layer, a medium-grained asphalt concrete intermediate layer, a coarse-grained asphalt macadam bottom layer, a water-based road polymer-stabilized macadam base course, and a graded macadam cushion / subgrade. S200: Collect multi-dimensional spatial feature data of each province, construct a spatial weight matrix W based on the k-nearest neighbor method, and calculate the stone lithology correction coefficient K_stone and the climate-precipitation coupling correction coefficient β_couple; S300: Establish a WQY spatially suppressed constrained gradient booster model, whose loss function simultaneously contains a mean squared error term and a spatial autocorrelation regularization term γ·Σwᵢⱼ·(F(xᵢ)-F(xⱼ)). 2 And the dose soft constraint term λ_dose·Σψ(dose_i), where the spatial constraint pseudo residual formula is rᵢ = yᵢ - (1+γ)·Fᵢ + γ·F_neigh_i, where F_neigh_i is the weighted prediction mean of adjacent spatial units; S400: Using the trained WQY model as a surrogate model, combined with the NSGA-II multi-objective optimization algorithm, polymer dosage is taken as the core decision variable, and the three objective functions of life-cycle carbon emissions, life-cycle cost and polymer replacement rate are optimized at the same time. S500: Based on a five-climate zoning system and six application scenario-based design modules, it outputs zoning and scenario-based design guidelines and GIS thematic maps.

[0008] As a further aspect of the present invention: the WQY loss function in step S300 is defined as: L_WQY^total(F) = Σ(1 / 2)(yᵢ-F(xᵢ)) 2 + (γ / 2)·Σwᵢⱼ·(F(xᵢ)-F(xⱼ)) 2 + λ_dose·Σψ(dose_i), where γ∈(0.1~0.5) is the spatial inhibition intensity coefficient, λ_dose∈(0.01~0.1) is the dose soft constraint penalty coefficient, and ψ(dose_i)=max(0, dose_i-d_max) 2 +max(0, d_min-dose_i) 2 d_min=0.50% and d_max=0.60% are the compliant dose boundaries.

[0009] As a further aspect of this invention: the recommended dosage range and base layer thickness parameters for the five climate design zones are as follows: Zone 1, severe cold and arid: 0.50%~0.60%, 200mm; Zone 2, mild and humid: 0.50%~0.60%, 180mm; Zone 3, hot and rainy: 0.50%~0.60%, 180mm; Zone 4, high altitude freeze-thaw: 0.50%~0.60%, 200mm; Zone 5, general areas: 0.50%~0.60%, 180mm. The standard dosage for all zones is uniformly 0.50%, with a maximum dosage of 0.60%.

[0010] The formula for calculating K_stone is: K_stone = 1.15 - 0.002×CV - 0.005×needle and flake content, with a value range of 0.90~1.15; the formula for calculating β_couple is: β_couple = 1.0 + 0.08×(P_ann-800) / 1000 +0.10×(N_FT-50) / 100, with a value range of 0.88~1.20; the formula for dose comprehensive correction is: dose_final = dose_optimal × K_stone × β_couple, and the final dose is constrained within the compliant range by max(0.50, min(0.60, dose_final)).

[0011] The six application scenario-based design modules cover highway interchanges, heavy-duty mining roads, secondary highways, urban arterial roads, expressway mainlines, and rural roads. Each scenario independently provides a recommended range for polymer layer thickness and a dose adjustment coefficient, specifically: 200~220mm / 1.00 for highway interchanges; 180~220mm / 1.05 for heavy-duty mining roads; 150~180mm / 0.98 for secondary highways; 150~200mm / 1.00 for urban arterial roads; 180~220mm / 1.02 for expressway mainlines; and 150~180mm / 0.95 for rural roads.

[0012] The three objective functions of NSGA-II in step (4) are: minimizing life-cycle carbon emissions, minimizing life-cycle cost, and maximizing polymer substitution rate. After optimization, the optimal compromise solution is determined by the TOPSIS approximation ideal solution ranking method. The weight configuration of the three objectives is: life-cycle carbon emissions weight 0.4, life-cycle cost weight 0.35, and polymer substitution rate weight 0.25.

[0013] Initial construction cost comparison (polymer base layer approximately 65 yuan / cm·m) 2 Compared to traditional water-stabilized layers, which cost approximately 45 yuan / cm·m 2 Comparison of annual maintenance costs (polymer, no more than 3% / year; traditional, about 10% / year); Comparison of major and medium repair costs (polymer, 12-15 years; traditional, 8-10 years); Comparison of carbon tax costs (polymer, about 0.125 yuan / m³). 2 Traditionally around 4.99 yuan / m 2 (Calculated with a design base period of 30 years and a discount rate of 8% per year).

[0014] The hyperparameters were determined through 5-fold spatial cross-validation and grid search, and the final optimized parameters were: γ=0.2, number of trees M=200, learning rate η=0.05, and λ_dose=0.05; WQY model performance validation: carbon emission prediction R 2=0.918, cost forecast R 2 =0.895, and the predicted residual Morans I decreased from 0.42 in the conventional GBDT to 0.05.

[0015] The method also includes an optional BIM+GIS digital delivery step, which outputs road structure layer information in IFC4.0 format and zoning boundaries and recommended design parameters in GeoJSON format.

[0016] As a further aspect of the present invention, the model also includes a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements various functions of the algorithm model.

[0017] Compared with the prior art, the beneficial effects of the present invention are: Scenario-based design: Through customized design for six typical application scenarios, the accuracy of polymer layer thickness recommendation is improved by more than 30%, which can effectively avoid the problems of over-design or under-design. Precise dosage recommendation: Based on a comprehensive analysis of factors such as climate zone, stone type, and traffic level, the polymer dosage is precisely recommended with a dosage recommendation range accurate to the 0.01% level; Life cycle optimization: Taking the Dunshang Interchange of S40 Chizhou Expressway as an example, after the optimization design using this model, the total thickness of the pavement structure is reduced from 74cm to 61cm, a reduction of 13cm, saving 15-20% in direct construction costs and 20-25% in life cycle costs; Low-carbon and environmental benefits: According to Ergomax assessment and S40 project calculations, using polymer-stabilized crushed stone base course reduces carbon emissions by about 97.5% compared to traditional water-stabilized base course (ratio of about 1:40). A single project can reduce carbon emissions by about 4,645 tons, which is equivalent to carbon asset revenue of about RMB269,400 (calculated at a carbon price of RMB58 per ton). Durability and reliability: Engineering case studies show that the polymer-stabilized crushed stone base course designed using this model has been in service for 13 years (S211 Provincial Highway in Ya'an, Sichuan, constructed in 2009), without potholes, subsidence, water damage, ruts, or reflective cracking. Its design life can reach 30 years, and the major and minor repair cycle is extended to 12-15 years. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the structure of a polymer road structure design algorithm model for water-based roads according to the present invention; Figure 2 This is a schematic diagram of the logical relationships and data flow of each module in the design algorithm model of a water-based road polymer road structure of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1: East Extension of Kashi Road, a main urban road in Urumqi (1) Climate zoning: The project is located in Urumqi and is classified as Zone 1 (severe cold and arid zone); (2) Application scenario: Urban main roads, with a design axle spacing of 1×10 7 Each dose adjustment factor is 1.00; (3) Stone parameters: local crushed stone crushing value CV=22%, needle-like and flaky content 8%; (4) K_stone calculation: K_stone = 1.15 - 0.002×22 - 0.005×8 = 1.066; (5) β_couple calculation: β_couple = 1.0 + 0.08×(300-800) / 1000 + 0.10×(90-50) / 100 = 1.000; (6) The WQY model outputs the optimal dose: dose_optimal = 0.50% (standard dose); (7) Overall correction: dose_final = 0.50% × 1.066 × 1.000 = 0.533%; (8) Recommended structure: surface layer 4+6+8cm, base layer thickness 200mm; (9) LCCA results: Compared with the traditional 36cm water-stabilized layer scheme, the total life cycle cost is reduced by about 22%, and carbon emissions are reduced by about 94.7%.

[0021] Actual engineering verification: The project has been completed for 5 years and is in good working order, with a thickness reduction of 50mm.

[0022] Example 2: Chizhou section of S40 Ningguo to Zongyang Expressway The project is located in Chizhou City, Anhui Province, and the application scenario is a highway interchange with a heavy traffic level.

[0023] The original design used a pavement structure of three layers of asphalt, two layers of water-stabilized pavement, and one layer of crushed stone, with a total thickness of 74cm. The optimized design using the model of this invention is shown in the table below: After optimization using the model of this invention, the total thickness of the pavement structure was reduced from 74cm to 61cm, a reduction of 13cm (17.6%), achieving a significant structural optimization effect. Upon acceptance, all performance indicators of the optimized scheme met the design requirements: compaction degree >98%, CBR >200%, smoothness <1.2cm, and residual moisture content <1%.

[0024] Example 3: Yuba Road, Xiqu District, Panzhihua City, Sichuan Province (Heavy-duty mining area road) Located in Xiqu District, Panzhihua City, Sichuan Province, the project is a heavy-duty road for transporting iron ore. It is 2.82 km long, 8 m wide, with a maximum longitudinal slope of 9% and a minimum turning radius of 31 m. The application scenario is a heavy-duty mining area road, and the traffic classification is heavy traffic (Highway-I).

[0025] After adopting the model design of this invention, the polymer-stabilized crushed stone layer thickness is 220mm, the dosage is 0.55%, the total thickness of the asphalt layer is 100mm, the lower water-stabilized slag layer is 150mm, and the total thickness of the pavement structure is 470mm. Compared with the original design (730mm), this is a reduction of 260mm, saving approximately 20% in material costs.

[0026] The project was completed and opened to traffic in August 2014. As of 2026, it has been in use for about 8 years. The road surface is in good condition after inspection, with no potholes, subsidence, water damage, ruts and reflective cracking. This verifies the reliability and durability of the model of this invention in the application of heavy-duty mining area roads.

[0027] Please see Figures 1-2 The polymer road structure design algorithm model provided by this invention adopts a modular architecture design, with each module communicating through a standardized data interface. The main model class acts as the core scheduler, responsible for coordinating the calls and data flow of each functional module.

[0028] 1. Data Class Definition The model defines the following core data classes: Climate zones are enumerated: five climate zones are defined: Zone 1 (severe cold and arid), Zone 2 (mild and humid), Zone 3 (hot and rainy), Zone 4 (high altitude freeze-thaw), and Zone 5 (general areas); Traffic level enumeration: Defines four levels: extra-heavy and above traffic, heavy traffic, medium traffic, and light traffic; Highway Class Enumeration: Defines two classes: expressways and Class I highways, and Class II and lower highways; Grading type enumeration: Four gradation types are defined: T-19, T-25, T-25(Q), and T-30. Application scenarios are enumerated: six application scenarios are defined, namely highway interchanges, heavy-duty mining area roads, secondary highways, urban arterial roads, expressway main lines, and rural roads.

[0029] Polymer technical requirements: Defines the technical specifications for water-based road polymers, including pH range (8-9), solid content range (28-38%), viscosity range (50-100 seconds), etc. Strength requirements: Minimum requirements for CBR and penetration strength (RT) are defined for different highway grades and traffic grades.

[0030] 2. Core Functional Modules Scenario-based design module: This module is one of the core innovations of this invention. A scenario-based recommendation database has been established for six typical application scenarios.

[0031] The core method of this module is design_for_scenario(scenario, traffic_level, climate_zone, highway_class), which automatically returns a recommended structural design scheme based on the input application scenario and traffic level.

[0032] Engineering Case Database: This module includes six typical engineering case studies, forming a structured database of engineering case studies. Each case study contains four main categories of data: basic project information, traditional structural design, polymer structural design, performance characteristics, and key experiences. Three query interfaces are provided: get_case(), get_cases_by_scenario(), and get_recommended_structure().

[0033] Precise Dose Gradient Design Module: The core method of this module is `recommend_dosage()`, which implements a multi-factor comprehensive recommendation algorithm for polymer dosage. The dosage recommendation formula is: `d_recommend = d_base × γ_stone × γ_couple × γ_traffic × γ_scenario`. The final dosage is limited to the range of 0.50%-0.60%.

[0034] Life cycle cost analysis module: This module implements Life Cycle Cost Analysis (LCCA) functionality, considering initial construction costs, discounted annual maintenance costs, discounted major and minor repair costs, and carbon tax costs. The discounting calculation uses the standard discount factor formula: .

[0035] Carbon emission refined management module: This module establishes a carbon emission factor database based on third-party assessment data from Ergomax. The carbon emission factor for the polymer method is 1.89 kg CO2 / m³. 2 The carbon emission factor of the cementing process is 14.23 kg CO2 / m³. 2 The ratio is approximately 1:7.5.

[0036] Multilayer Elastic System Mechanical Analysis Module: This module implements pavement structure mechanical analysis based on multi-layer elastic system theory, and integrates calculation methods for six core verification indicators: pavement surface acceptance deflection, asphalt layer permanent deformation, asphalt layer fatigue cracking, polymer base fatigue cracking, penetration strength, and subgrade compressive strain.

[0037] The engineering case database must include at least the following engineering cases: When building the precise design module for the metrological gradient, follow these steps: Step 1: Construct a five-layer standard water-based road polymer road structure system Based on the JTG D50-2017 standard, a five-layer standard composite structure system is constructed, from top to bottom as follows: (1) Fine-grained asphalt concrete surface layer (AC-13): 4cm thick; (2) Medium-grained asphalt concrete intermediate layer (AC-20): thickness 5~6cm; (3) Coarse-grained asphalt macadam subbase (ATB-25): 8~10cm thick; (4) Polymer-stabilized crushed stone base course for water-based roads: The polymer dosage is a design variable, with a standard dosage of 0.50% and a maximum of 0.60%, and a thickness of 150~180mm; (5) Graded crushed stone cushion layer / subgrade: The thickness is determined based on the compressive strain calculation results of the subgrade.

[0038] Step 2: Collect spatial feature data and construct a spatial weight matrix Multi-dimensional spatial feature data from various provinces were collected, comprising over 30 parameters across 7 categories. The spatial weight matrix W was constructed using the k-nearest neighbor method (k=5), with the weight coefficients of neighboring provinces normalized.

[0039] Step 3: Establish a WQY space-suppression-constrained gradient booster model (I) Definition of WQY loss function The core innovation of the WQY model lies in the simultaneous introduction of a spatial autocorrelation regularization term and a dose soft constraint term into the loss function. The loss function is defined as: + λ_dose·Σᵢ ψ(dose_i) in: ① The first item is the mean square error loss (standard monitoring loss); ② The second term is the spatial autocorrelation regularization term, γ>0 is the spatial suppression strength coefficient (preferred range 0.1~0.5), and wᵢⱼ is the element of the spatial weight matrix; ③ The third term is the dose soft constraint term, where λ_dose>0 is the penalty coefficient (preferred range 0.01~0.1). The dose soft constraint function is: ψ(dose_i) = max(0, dose_i - d_max) 2 + max(0, d_min - dose_i) 2 d_min=0.50% and d_max=0.60% are the compliant dose boundaries.

[0040] (II) Derivation of the formula for spatially constrained pseudo-residuals The formula for spatially constrained pseudo-residuals is: rᵢ = yᵢ - (1+γ)·Fᵢ + γ·F_neigh_i Where F_neigh_i is the weighted prediction mean of neighboring spatial cells. The derivation process is as follows: The gradient descent update of the WQY model is based on the partial derivative of the space-constrained loss function. Taking the partial derivative of the space regularization term with respect to F(xᵢ): ∂ / ∂F(xᵢ) [(γ / 2)·Σⱼ wᵢⱼ·(F(xᵢ)-F(xⱼ)) 2 ] = γ·Σⱼ wᵢⱼ·(F(xᵢ)-F(xⱼ)) Introducing this into the gradient boosting framework, the space-constrained pseudo-residual is corrected as follows: rᵢ = yᵢ - (1+γ)·F_{t-1}(xᵢ) + γ·F_neigh_{t-1}(xᵢ) (III) Algorithm Convergence Analysis The convergence of the WQY model can be demonstrated from the following three aspects: ① Convexity of loss function: When the base learner is a decision tree stump, the WQY loss function is convex with respect to F and has a global optimum. ② Gradient descent convergence: The spatially constrained pseudo-residual formula ensures that the gradient direction in each iteration points to the direction of the fastest decrease in the loss function. According to convex optimization theory, the gradient descent method must converge. ③ Verification of spatial suppression effect: After five-fold spatial cross-validation of 31 provincial-level administrative units, the Morans I of the WQY model prediction residual decreased from 0.42 in the traditional GBDT to 0.05, indicating that spatial autocorrelation was effectively suppressed.

[0041] (iv) Performance verification and comparative experiment of WQY model Experimental data source: Samples from 31 provincial-level administrative units, using 5-fold spatial cross-validation. Results show that the WQY model improves carbon emission prediction and cost prediction by 15.2% and 16.4% respectively, and reduces the prediction residual Morans I by 88%.

[0042] Step 4: Differentiated Design and Scenario-Based Design Module for Five Climate Zones (I) Differentiated Design for Five Climate Zones The standard dose for each zone is uniformly set at 0.50%, and the maximum dose is 0.60%. Differentiated and precise design for each zone is achieved through K_stone and β_couple correction coefficients.

[0043] (II) Scenario-based design module Step 5: Integration with NSGA-II multi-objective optimization algorithm Using the trained WQY model as a surrogate model and combining it with the NSGA-II multi-objective optimization algorithm, the polymer dosage is taken as the core decision variable, and the optimal polymer dosage and structural parameters for each province are solved under the six mechanical verification constraints specified in JTG D50-2017.

[0044] Three optimization objective functions were defined: ① Minimize lifetime carbon emissions; ② Minimize lifetime cost (LCCA comprehensive accounting); ③ Maximize polymer substitution rate. After optimization using NSGA-II, the TOPSIS approximation method was employed to determine the optimal compromise solution, with the following weighting: carbon emissions 0.4, cost 0.35, and substitution rate 0.25.

[0045] Step 6: Life Cycle Cost Analysis (LCCA) A life-cycle cost comparison analysis was conducted between the polymer solution and the traditional cement-stabilized crushed stone solution, with a design baseline of 30 years and a discount rate of 8% per year, covering: initial construction cost (polymer base layer approximately 65 RMB / cm²). 2 Traditional water-stabilized layers cost approximately 45 yuan / cm·m. 2 Annual maintenance cost discount (polymer no more than 3% / year, traditional about 10% / year); major and medium repair cost discount (polymer 12~15 years, traditional 8~10 years); carbon tax cost (polymer about 0.125 yuan / m³). 2 Traditionally around 4.99 yuan / m 2 ).

[0046] Step 7: Output the zoning design guide and GIS thematic map Based on the NSGA-II optimization results, and by combining K_stone correction and β_couple correction, a national recommended map of polymer road structure design zones (five zones) and a series of GIS thematic maps, as well as customized design parameter manuals for various application scenarios, are generated.

[0047] Dose comprehensive correction system Formula for calculating the stone lithology correction factor K_stone: K_stone = 1.15 - 0.002×CV - 0.005×needle-like content (value range 0.90~1.15) Verified by 48 sets of typical engineering gravel lithology test data, the CBR prediction error after K_stone correction was reduced from ±18% to ±7%.

[0048] Formula for calculating the climate-precipitation coupling correction coefficient β_couple: β_couple = 1.0 + 0.08×(P_ann-800) / 1000 + 0.10×(N_FT-50) / 100, with a value range of 0.88~1.20. The comprehensive dose correction formula is: dose_final = dose_optimal × K_stone × β_couple. The final dose is constrained within the compliant range by max(0.50, min(0.60, dose_final)).

[0049] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A design algorithm model for polymer road structures used in water-based roads, comprising an application scenario module, a case database, a gradient design module, a cost analysis module, a management module, and a mechanical analysis module, characterized in that: The application scenario-based design module is used to provide customized polymer-stabilized crushed stone base structure design schemes based on one of the following application scenarios and traffic levels: highway interchanges, heavy-duty mining area roads, secondary highways, urban arterial roads, expressway main lines, and rural roads. The engineering case database is used to store and manage the structural design parameters and performance data of typical engineering cases from both domestic and international sources. The dosage gradient precision design module is used to comprehensively calculate and recommend the optimal polymer dosage based on factors such as climate zone, stone type, traffic level and application scenario; The life-cycle cost analysis module is used to comprehensively calculate the present value of the initial construction cost, maintenance cost, major and minor repair cost, and carbon tax cost for both polymer and traditional solutions. The carbon emission refined management module is used to calculate and compare the carbon emissions of polymer solutions and traditional solutions based on the carbon emission factor database; The multi-layer elastic system mechanical analysis module is used to calculate and verify six core indicators: road surface acceptance deflection, permanent deformation of asphalt layer, fatigue cracking of asphalt layer, fatigue cracking of polymer base course, penetration strength, and roadbed compressive strain.

2. The design algorithm model for a polymer road structure for water-based roads according to claim 1, characterized in that: The precise dose gradient design module uses the following dose calculation formula: d_recommend = d_base × γ_stone × γ_couple × γ_traffic × γ_scenario, where d_base is the baseline dose of 0.50%, γ_stone is the stone correction factor, γ_couple is the climate coupling factor, γ_traffic is the traffic adjustment factor, and γ_scenario is the scene adjustment factor. The final dose is limited to the range of 0.50%-0.60%. The precise dose gradient design module establishes a unified response relationship between polymer dose and CBR intensity and penetration intensity: CBR(d) = 100 + d × 300; RT(d) = 0.35 + d × 0.

75.

3. The design algorithm model for a polymer road structure for water-based roads according to claim 1, characterized in that: The mechanical analysis module for the multilayer elastic system adopts a fatigue life correction formula for the polymer base layer based on the stress ratio theory: Where b_e=9.0 is the fatigue index.

4. The design algorithm model for a polymer road structure for water-based roads according to claim 1, characterized in that: The engineering case database contains at least six typical engineering cases, each of which includes basic project information, traditional structural design parameters, polymer structural design parameters, performance data, and key experience data.

5. The design algorithm model for a polymer road structure for water-based roads according to claim 1, characterized in that: The life-cycle cost analysis module uses the standard discount factor formula to calculate the present value of costs: .

6. The design algorithm model for a polymer road structure for water-based roads according to claim 1, characterized in that: The polymer stabilized aggregate method carbon emission factor of the carbon emission fine management module is 1.89 kg CO2 / m 2 The traditional cement stabilized aggregate method carbon emission factor is 14.23 kg CO2 / m 2 The ratio is about 1:7.

5.

7. The design algorithm model for a polymer road structure for water-based roads according to claim 1, characterized in that: The precise design module for the metrological gradient includes the following calculation steps: S100: Based on the JTG D50-2017 specification, a five-layer standard water-based road polymer road structure system is constructed, consisting of a fine-grained asphalt concrete surface layer, a medium-grained asphalt concrete intermediate layer, a coarse-grained asphalt macadam bottom layer, a water-based road polymer-stabilized macadam base course, and a graded macadam cushion / subgrade. S200: Collect multi-dimensional spatial feature data from all provinces in China, construct a spatial weight matrix W based on the k-nearest neighbor method, and calculate the stone lithology correction coefficient K_stone and the climate-precipitation coupling correction coefficient β_couple. S300: Establish the WQY spatially restrained gradient boosting model, whose loss function contains the mean square error term, the spatial autocorrelation regularization term γ·∑wᵢⱼ·(F(xᵢ)-F(xⱼ)) 2 and the dose soft constraint term λ_dose·∑ψ(dose_i), where the spatial constraint pseudo-residual formula is rᵢ = yᵢ - (1+γ)·Fᵢ + γ·F_neigh_i, F_neigh_i is the weighted prediction mean of the adjacent spatial unit; S400: Using the trained WQY model as a surrogate model, combined with the NSGA-II multi-objective optimization algorithm, polymer dosage is taken as the core decision variable, and the three objective functions of life-cycle carbon emissions, life-cycle cost and polymer replacement rate are optimized at the same time. S500: Based on a five-climate zoning system and six application scenario-based design modules, it outputs zoning and scenario-based design guidelines and GIS thematic maps.

8. The design algorithm model for a polymer road structure for water-based roads according to claim 7, characterized in that: The WQY loss function mentioned in step S300 is defined as: L_WQY^total(F) = Σ(1 / 2)(yᵢ-F(xᵢ)) 2 + (γ / 2)·Σwᵢⱼ·(F(xᵢ)-F(xⱼ)) 2 + λ_dose·Σψ(dose_i), where γ∈(0.1~0.5) is the spatial inhibition intensity coefficient, λ_dose∈(0.01~0.1) is the dose soft constraint penalty coefficient, and ψ(dose_i)=max(0, dose_i-d_max) 2 +max(0, d_min-dose_i) 2 d_min=0.50% and d_max=0.60% are the compliant dose boundaries.

Images