Medium-and-long-term planning decision-making method and system for highway network-level pavement maintenance
Through optimized data management and adaptive machine learning analysis, combined with the preprocessing and simulation methods of the adversarial generation network, the problems of large amount of data and high analysis difficulty in highway pavement maintenance planning are solved, and the scientific construction and maintenance planning of road network-level high-speed construction and maintenance planning are realized and accuracy meet medium- and long-term planning needs.
Patent Information
- Application Number
- CN202411036055.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-11-15
AI Technical Summary
The existing highway pavement maintenance planning decision model has a complex amount of data, high manual analysis is difficult and low accuracy, and it is impossible to effectively and automatically predict and meet medium- and long-term planning needs.
Data collection and storage management are used to optimize high-speed construction and maintenance data of road network level, and large data is analyzed using adaptive optimization machine learning methods, and weighted summing is constructed through road network maintenance planning data preprocessing and simulation methods based on adversarial generation network. Decision model, dynamic sorting maintenance order.
It effectively solves the problems of complex data and high difficulty in manual analysis, improves the scientificity and accuracy of road network-level high-speed construction and maintenance planning, and meets the scientific level of medium- and long-term planning decisions.
Smart Images

Figure CN118967103A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of highway planning and construction, and more specifically to a medium- and long-term planning decision-making method and system for highway network-level pavement maintenance. Background Art
[0002] With the increase of expressway operating mileage, the proportion of key sections has increased, the road age has generally become higher, the maintenance volume has begun to increase, and the maintenance management tasks have become increasingly prominent. It is urgent to build a medium- and long-term planning decision-making model for highway network-level pavement maintenance to optimize maintenance management decisions and provide a scientific basis for the implementation of maintenance projects.
[0003] The medium- and long-term planning decision model for highway network-level pavement maintenance refers to the use of a planning decision model to make data-driven decisions on regional provincial highway planning. It collects, analyzes, preprocesses, and simulates regional provincial highway construction and maintenance data, including operation management data, road section properties, technical standards, operating years, traffic volume, and traffic composition data, to make decisions on the sustainable development of regional provincial highways and optimize regional provincial highway construction and maintenance plans, thereby providing support for the sustainable development of regional provincial highways.
[0004] However, there are some technical problems in the existing highway pavement maintenance planning decision-making model. These problems make the storage and management of big data complicated and difficult, which in turn affects the efficiency and accuracy of subsequent big data analysis and the availability of regional provincial highway planning and construction tasks; based on the existing highway pavement maintenance planning decision-making model, the data volume is huge, the manual analysis is difficult and the accuracy is low, it is difficult to effectively predict automatically, the simulation decision is unreasonable, and it cannot meet the medium and long-term planning needs and other technical problems.
[0005] Therefore, it is an urgent problem for technical personnel in this field to provide a medium- and long-term planning decision-making method and system for highway network-level pavement maintenance that optimizes the allocation of maintenance funds, extends the service life of roads, improves resource utilization efficiency, and ensures the level of safe operation. Summary of the invention
[0006] In view of this, the present invention provides a method and system for medium- and long-term planning decision-making of highway road network-level pavement maintenance, which optimizes the problem of complex data in highway construction and maintenance at the highway network level by adopting data collection and storage management; adopts an adaptive optimization machine learning method to analyze highway big data at the highway network level, eliminating the defects of high difficulty and low accuracy of manual analysis; adopts a road network maintenance planning data preprocessing and road network maintenance planning simulation method based on adversarial generative network to ensure the scientific nature of highway construction and maintenance planning decision-making at the highway network level, and finally adopts a maintenance decision-making model to dynamically solve the maintenance sorting problem of existing highway networks by using a weighted summation method, which effectively promotes the scientific degree and accuracy of medium- and long-term planning decisions for highway maintenance in regional provinces.
[0007] In order to achieve the above object, the present invention adopts the following technical solution:
[0008] On the one hand, the present invention provides a method for medium- and long-term planning and decision-making of highway network-level pavement maintenance, comprising:
[0009] S1: Acquire original data of a regional highway network, pre-process the original data of the regional highway network, and store the data to obtain a regional highway network construction and maintenance data set;
[0010] S2: Analyze and evaluate the regional highway network according to the regional highway network construction and maintenance data set to obtain a highway network evaluation result;
[0011] S3: Obtaining road network maintenance planning data according to the regional highway network construction and maintenance data set and the highway network evaluation result;
[0012] S4: performing simulation and simulation evaluation according to the road network maintenance planning data to obtain a road network maintenance planning simulation evaluation result;
[0013] S5: Making a road network high-speed maintenance planning decision based on the road network maintenance planning simulation evaluation results.
[0014] Preferably, the S1 specifically includes:
[0015] S101: Obtain the original data of the regional highway network and construct the original data set of the highway network;
[0016] S102: compressing the original data in the highway network original data set into blocks;
[0017] S103: using arithmetic logic coding to encode the compressed regional highway network original data set to create a data identifier, and performing structured processing according to the identifier, and aggregating the structured regional highway network original data set to obtain a regional highway network construction and maintenance data set;
[0018] S104: Screening the data in the regional highway network construction and maintenance data set according to the adaptability of the data, removing invalid data, and eliminating discrete and repeated data.
[0019] Preferably, the S104 specifically includes:
[0020] Adaptation detection is performed using the marked highway network data code to obtain a road network data entropy value table, and each entropy value with repetition and discreteness is listed according to the set entropy threshold MENT to screen the adaptation grouping function G, and the grouping function G is:
[0021]
[0022] In the formula, G() is the grouping function, x i is the i-th data column, i is the data column identifier, K n is the nth data set, K n+m is another data set, ENT(x i ) is the entropy value of the ith data column, MENT is the entropy threshold, and α is the threshold bias parameter.
[0023] Preferably, the S3 specifically includes:
[0024] S301: Constructing a road network maintenance planning data preprocessing analysis model;
[0025] S302: Analyze the data in the regional expressway network construction and maintenance data set and the expressway network evaluation results according to the road network maintenance planning data preprocessing analysis model to obtain road network maintenance planning data.
[0026] Preferably, the S301 specifically includes: weight allocation, training weak differentiators using a decision tree algorithm, constructing a loss function, updating weights, and weighted voting.
[0027] Preferably, S4 includes:
[0028] Constructing a discriminator and a generator, and loss functions corresponding to the discriminator and the generator respectively;
[0029] The discriminator and the generator are trained respectively using corresponding loss functions to obtain a trained discriminator and generator;
[0030] Use the trained discriminator to assign probability values to the preset regional highway construction and maintenance planning simulation images;
[0031] The trained generator enhances the data set of the regional highway construction and maintenance planning simulation image assigned with probability values according to the road network maintenance planning data to generate a predicted simulation image;
[0032] Acquire a set of simulation images according to the simulation images predicted by the generator;
[0033] The maintenance plan is evaluated according to the simulation image set and the road network maintenance planning data to obtain a road network maintenance planning simulation evaluation result.
[0034] Preferably, the S5 specifically includes:
[0035] According to the simulation evaluation results of the road network maintenance planning, the road section properties, operating years and traffic volume of different road sections are obtained, and the maintenance type is determined according to the road section properties, operating years and traffic volume;
[0036] Determine the influencing factors of different road sections according to the simulation evaluation results of the road network maintenance planning, and assign weights according to the importance, contribution and controllability of the influencing factors to obtain a maintenance demand index;
[0037] According to the maintenance demand index combined with annual maintenance investment, maintenance countermeasure technology, maintenance management, and regional funds, the road network highway maintenance planning decision is obtained.
[0038] On the other hand, the present invention provides a highway road network-level pavement maintenance medium- and long-term planning decision system, including: a data collection and storage management module, a road network-level big data analysis and evaluation module, a road network maintenance planning data preprocessing module, a road network maintenance planning simulation module, and a road network high-speed maintenance planning decision module;
[0039] The data collection and storage management module is used to obtain the original data of the regional highway network, pre-process the original data of the regional highway network, and store it to obtain a regional highway network construction and maintenance data set;
[0040] A road network-level highway big data analysis and evaluation module is used to analyze and evaluate the regional highway network based on the regional highway network construction and maintenance data set to obtain a highway network evaluation result;
[0041] A road network maintenance planning data preprocessing module, used to obtain road network maintenance planning data according to the regional highway network construction and maintenance data set and the highway network evaluation result;
[0042] A road network maintenance planning simulation module is used to perform simulation and simulation evaluation based on the road network maintenance planning data to obtain a road network maintenance planning simulation evaluation result;
[0043] The road network highway maintenance planning decision module is used to make road network highway maintenance planning decisions based on the road network maintenance planning simulation evaluation results.
[0044] Preferably, the data collection and storage management module specifically includes:
[0045] A data block compression unit, used to obtain the original data of the regional highway network, construct an original data set of the highway network, and compress the original data in the original data set of the highway network in blocks;
[0046] The data identification unit is used to encode the compressed regional highway network original data set using arithmetic logic coding to create data identification, and perform structured processing according to the identification.
[0047] A data adaptation and screening unit, used to screen the data in the regional highway network construction and maintenance data set according to the adaptability of the data, remove invalid data, and eliminate discrete and repeated data;
[0048] The road network maintenance data summary storage unit is used to summarize the regional highway network original data set after structured processing to obtain the regional highway network construction and maintenance data set.
[0049] Preferably, the road network maintenance planning simulation module specifically includes:
[0050] A loss function construction unit, used to construct the discriminator loss function and the generator loss function respectively;
[0051] A training discriminator unit is used to construct a discriminator and train the discriminator based on the discriminator loss function;
[0052] A training generator unit, used for constructing a generator and training the generator based on the generator loss function;
[0053] A training and simulation model generation unit, which obtains a road network maintenance planning simulation model based on the trained discriminator and generator;
[0054] The road network maintenance planning simulation unit performs road network maintenance planning simulation based on the road network maintenance planning simulation model to obtain a simulation image set, and evaluates the maintenance plan based on the simulation image set and the road network maintenance planning data to obtain a road network maintenance planning simulation evaluation result.
[0055] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a method and system for medium- and long-term planning and decision-making of highway network-level pavement maintenance, which has the following beneficial effects:
[0056] (1) Aiming at the problem of large amount of data in the existing highway pavement maintenance planning decision model, the present invention establishes an optimized data warehouse, compresses and partitions the original data of the regional highway network, identifies and summarizes the data, and screens and adapts the data to obtain a regional highway network construction and maintenance data set, thereby solving the problem of large amount of highway construction and maintenance data at the road network level;
[0057] (2) In view of the problem that manual analysis is difficult and has low accuracy in the existing highway pavement maintenance planning decision model, the present invention adopts an adaptive optimization machine learning method to analyze the highway big data at the road network level, and uses machine learning to extract characteristic data from the highway construction and maintenance historical data at the road network level, laying a foundation for the accuracy and adaptability of the next step of data preprocessing and road network maintenance planning simulation;
[0058] (3) In view of the difficulty in effective automatic prediction and unreasonable simulation decision-making in the existing highway pavement maintenance planning decision-making model, the present invention adopts a road network maintenance planning data preprocessing and road network maintenance planning simulation method based on a generative adversarial network to ensure the scientificity and rationality of the road network-level highway construction and maintenance planning decision-making;
[0059] (4) In order to address the problem that the existing highway pavement maintenance planning decision model cannot meet the medium- and long-term planning needs, the present invention provides a maintenance decision model for dynamically sorting the existing highway maintenance planning sequence using a weighted summation method, thereby effectively meeting the scientific level of medium- and long-term planning decisions for highway maintenance in regional provinces. BRIEF DESCRIPTION OF THE DRAWINGS
[0060] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.
[0061] Figure 1 A schematic diagram of a process flow of a medium- and long-term planning decision-making method for highway network-level pavement maintenance provided by the present invention;
[0062] Figure 2 It is a schematic diagram of the process of S1;
[0063] Figure 3 It is a schematic diagram of the process of S3;
[0064] Figure 4 is a schematic diagram of the process of S4;
[0065] Figure 5 This is a schematic diagram of the maintenance planning scope model for S4;
[0066] Figure 6 is a schematic diagram of the process of S5;
[0067] Figure 7 A schematic diagram of the structure of a medium- and long-term planning and decision-making system for highway network-level pavement maintenance provided by the present invention;
[0068] Figure 8It is a structural diagram of the data collection and storage management module;
[0069] Fig. 9 Schematic diagram of the structure of the mini-module for road network maintenance planning. DETAILED DESCRIPTION
[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0071] The embodiment of the present invention discloses a method for making mid- and long-term planning decisions for highway network-level pavement maintenance. Figure 1 As shown, including:
[0072] S1: Acquire the original data of the regional highway network, pre-process the original data of the regional highway network, and store it to obtain a regional highway network construction and maintenance data set.
[0073] Specifically, the original data is collected and sorted through the regional provincial highway network construction and maintenance platform to obtain the regional highway original data set, where the regional highway original data set includes operation management data, road section properties, technical standards, operating years, traffic volume and traffic composition. Data storage management optimization includes establishing an optimized data warehouse, storing and managing the regional highway original data set, and obtaining the regional highway encrypted data set and key.
[0074] S2: Analyze and evaluate the regional highway network based on the regional highway network construction and maintenance dataset to obtain the highway network evaluation results.
[0075] Specifically, the road network profile is extracted from the regional highway network construction and maintenance data set for analysis, where the road network profile includes basic information such as operation management, road section nature, technical standards, operating years, traffic volume and traffic composition. The regional highway network evaluation is to summarize the maintenance history of the road network in various directions in the past three years, conduct a technical evaluation of the current road performance index, sort out the existing major problems, conduct a regional provincial highway planning evaluation, and pre-process the planning data based on the evaluation results.
[0076] S3: Obtain road network maintenance planning data based on the regional highway network construction and maintenance data set and highway network evaluation results; Screen the highways that meet the requirements of road network operation life by using the conditional function M to construct a maintenance planning scope model. The formula M is:
[0077]
[0078] Where M() is the conditional function of the expressways not included in the planning scope, A i is the ith expressway, i is the data identifier of the expressway, Z is the regional expressway network construction and maintenance data set, P() is the operating life function, and q() is the treatment life function;
[0079] And obtain highway network evaluation results to obtain road network maintenance planning data.
[0080] Among them, the road network maintenance planning data is based on the regional provincial highway big data basic data set and the regional highway maintenance planning geographic satellite image set. The machine adaptive optimization method is used and the loss function model is constructed to obtain the road network maintenance planning data preprocessing analysis model. According to the model, the road network-level highway big data set is analyzed to optimize the road network maintenance planning data.
[0081] S4: Perform simulation and simulation evaluation based on the road network maintenance planning data to obtain the road network maintenance planning simulation evaluation results.
[0082] Road network maintenance planning simulation, conduct simulation evaluation on the road network maintenance planning scope model, screen out expressways whose road network operating life meets the requirements based on the road network maintenance planning data, use the provincial expressway network platform to build a road network maintenance planning simulation information module to conduct road network maintenance planning simulation.
[0083] S5: Make decisions on the road network maintenance planning based on the road network maintenance planning simulation evaluation results.
[0084] Among them, the road network expressway maintenance planning decision is to divide the road network maintenance units according to the established road network maintenance planning simulation information module, and then make decisions on maintenance types, select maintenance countermeasures, build a maintenance demand ranking model, and optimize the dynamic allocation model of maintenance planning.
[0085] Further, such as Figure 2 As shown, S1 specifically includes:
[0086] S101: Acquire the original data of the regional highway network and construct the original data set of the highway network.
[0087] S102: compressing the original data in the highway network original data set into blocks;
[0088] S103: using arithmetic logic coding to encode the compressed regional highway network original data set to create a data identifier, and performing structured processing according to the identifier, and aggregating the structured regional highway network original data set to obtain a regional highway network construction and maintenance data set;
[0089] S104: Screen the data in the regional highway network construction and maintenance data set according to the adaptability of the data, remove invalid data, and eliminate discrete and repeated data. Adaptability is to select appropriate reference data based on regional highways. Some highways have "provincial inspection databases", some highways use "national inspection databases", and some highways refer to the recent "self-inspection database". Different databases have different time, inspection items, and inspection standards. According to the different reference databases and maintenance plans, select appropriate data and eliminate discrete and repeated data.
[0090] Furthermore, S104 specifically includes:
[0091] The marked highway network data code is used for adaptation detection to obtain the road network data entropy value table. The highway network data code refers to the highway maintenance projects of different road networks. After marking, different maintenance categories and maintenance parts of the road section are given certain codes for standardized storage.
[0092] Among them, the road network data entropy value table contains the entropy values of operation management data, road section properties, technical standards, operating years, traffic volume and traffic composition. Each repeated and discrete entropy value is listed according to the set entropy threshold MENT to screen the adaptive grouping function G. The grouping function G is:
[0093]
[0094] In the formula, G() is the grouping function, x i is the i-th data column, i is the data column identifier, K n is the nth data set, K n+m is another data set, ENT(x i ) is the entropy value of the ith data column, MENT is the entropy threshold, and α is the threshold bias parameter.
[0095] like Figure 3 As shown, in another embodiment, S3 specifically includes:
[0096] S301: Constructing a road network maintenance planning data preprocessing analysis model;
[0097] S302: Analyze the data in the regional expressway network construction and maintenance data set according to the road network maintenance planning data preprocessing analysis model to obtain road network maintenance planning data.
[0098] Specifically, S301 includes: weight allocation, decision tree algorithm training weak differentiators, building loss function, updating weights, and weighted voting.
[0099] The specific weight allocation is to randomly select 90% of the data from the regional provincial highway construction and maintenance data set as the training sample set, and assign equal weights to each training sample in the selected training sample set based on the regional provincial highway big data summary data table; use the decision tree algorithm to train the weak classifier H for simple preliminary classification of the data; the loss function L is used to calculate the average loss to optimize the training, and the function L is:
[0100]
[0101] Where L is the loss function, t i is the true identity of the i-th sample, y i is the prediction flag of the weak classifier H, n is the total number of samples, and N is the total weight of the samples;
[0102] Furthermore, in step S3, the updated weights are used for model accumulation training, and the value of the sample after the updated weights is calculated by updating the weight parameters. The calculation formula is:
[0103]
[0104] In the formula, θ i is the updated weight value, L is the loss function, and N is the total weight number of samples.
[0105] Furthermore, in step S3, weighted voting is used to determine the final prediction result, specifically, updating the weight value, using the weak classifier H for preliminary classification, and obtaining the data prediction result.
[0106] like Figure 4 and Figure 5 As shown, in another implementation, S4 includes:
[0107] The discriminator and generator, as well as the loss functions corresponding to the discriminator and generator, are constructed separately; the discriminator is constructed using a convolutional neural network and a rectified linear unit activated convolutional layer.
[0108] The discriminator and generator are trained using the corresponding loss functions to obtain the trained discriminator and generator. Specifically, the adversarial loss function LGRA is used to optimize the quality of the regional provincial highway construction and maintenance planning image obtained by simulation. The formula is:
[0109]
[0110] Where LGRA is the generator adversarial loss function, x r is the real data, x f is to generate data, is the mathematical expectation of the real data, is the mathematical expectation of the generated data, D RA() is the discriminator network identification function;
[0111] The trained discriminator is used to assign probability values to the preset regional highway construction and maintenance planning simulation images; the preset highway construction and maintenance planning simulation map is compiled based on the planning target files of the provincial and ministerial documents and the owner units such as the Transportation Investment Group, and the simulation map is generated through image processing files and mapgis. The default simulation map has been generated here, and the main process is to optimize it.
[0112] The trained generator enhances the data set of regional highway construction and maintenance planning simulation images assigned probability values based on the road network maintenance planning data to generate predicted simulation images; the trained generator enhances the image according to the data, improves the combination of data and image, and performs accurate and complete simulation under the premise of ensuring the visibility of the image data.
[0113] According to the simulated image predicted by the generator, a set of simulated images is obtained;
[0114] The maintenance plan is evaluated based on the simulation image set and the road network maintenance planning data to obtain the road network maintenance planning simulation evaluation results.
[0115] Further, such as Figure 6 As shown, S5 specifically includes:
[0116] According to the simulation and evaluation results of the road network maintenance plan, the road section nature, operating life and traffic volume of different road sections are obtained, and the maintenance type is determined according to the road section nature, operating life and traffic volume.
[0117] Among them, the road section nature is divided into: particularly important national trunk line, important national trunk line, important provincial trunk line, other national highways, other provincial highways, and its importance coefficient is set to ZY, and its weight is w1;
[0118] The operating years are classified according to the empirical law of asphalt pavement performance attenuation, and are divided into 5-8 years, 9-12 years, 13-15 years, and >15 years. The importance coefficient is set to NX, and the weight is w2.
[0119] The traffic volume is converted to the annual average daily equivalent traffic volume of a single lane, which is classified into five levels: <5000, 5000-8000, 8000-12000, 12000-20000, and >20000. The importance coefficient is set to JT and the weight is w3;
[0120] The pavement maintenance type decision model is established as follows:
[0121] LX=ZY×NX×JT
[0122]
[0123] Where:
[0124] ZY—importance coefficient (indicates: road section importance category);
[0125] NX—operation life coefficient (indicates: operation life);
[0126] JT—traffic volume coefficient (expressed as: single lane equivalent traffic volume);
[0127] LX—Maintenance type decision value (maintenance type coefficient).
[0128] According to the simulation and evaluation results of road network maintenance planning, the influencing factors of different road sections are determined, and weights are allocated according to the importance, contribution and controllability of the influencing factors to obtain the maintenance demand index.
[0129] The maintenance demand ranking model ranks maintenance needs by allocating weights based on the importance, contribution, controllability and other factors of each influencing factor of the road section.
[0130] The technical condition mainly considers the road surface damage PCI, and other indicators such as flatness RQI, rutting RDI and anti-skid SRI are considered as auxiliary indicators, with a weight of w4; bridges and tunnels refer to the latest inspection report and are predicted based on the changes in technical conditions in recent years, with a weight of w5;
[0131] The PCI formula for road damage is:
[0132]
[0133] In the formula, w i is the weight of the i-th type of pavement damage, A i is the area of the i-th type of road damage (in m 2 ), A is the road surface detection area.
[0134] The flatness RQI formula is:
[0135]
[0136] Where IRI is the international roughness index measured on site (unit: m / km);
[0137] The rutting RDI formula is:
[0138]
[0139] Where H is the rutting depth (unit: mm);
[0140] The anti-slip SRI formula is:
[0141]
[0142] Where SFC is the lateral force coefficient;
[0143] The maintenance demand ranking model uses the weighted sum ranking method to rank the maintenance demand based on the four factors of regional highway network section nature, operating years, traffic volume, and technical conditions. The weighted sum ranking method formula is:
[0144] S i =W1SN i +W2ST i +W3SP i +W4SI i
[0145] Where: W1, W2, W3, W4, W5—are the corresponding weights of each evaluation factor;
[0146] S—comprehensive score;
[0147] SN—road section nature score;
[0148] ST—Years of Operation Score;
[0149] SP—traffic volume score;
[0150] SI—bridge and tunnel technical condition score.
[0151] According to the importance coefficient of maintenance demand combined with annual maintenance investment, maintenance strategy technology, maintenance management and regional funds, the highway maintenance planning decision of the road network is obtained.
[0152] On the other hand, the present invention provides a medium- and long-term planning decision system for highway network-level pavement maintenance, such as Figure 7 As shown, it includes: data collection and storage management module, road network level big data analysis and evaluation module, road network maintenance planning data preprocessing module, road network maintenance planning simulation module, road network high-speed maintenance planning decision module;
[0153] The data collection and storage management module is used to obtain the original data of the regional highway network, pre-process the original data of the regional highway network, and store it to obtain the regional highway network construction and maintenance data set;
[0154] The road network-level highway big data analysis and evaluation module is used to analyze and evaluate the regional highway network based on the regional highway network construction and maintenance data set to obtain the highway network evaluation results;
[0155] The road network maintenance planning data preprocessing module is used to obtain the road network maintenance planning data based on the regional highway network construction and maintenance data set and the highway network evaluation results;
[0156] The road network maintenance planning simulation module is used to perform simulation and simulation evaluation based on the road network maintenance planning data to obtain the road network maintenance planning simulation evaluation results;
[0157] The road network and highway maintenance planning decision-making module is used to make road network and highway maintenance planning decisions based on the road network maintenance planning simulation evaluation results.
[0158] like Figure 8 As shown, the data collection and storage management module specifically includes:
[0159] A data block compression unit is used to obtain the original data of the regional highway network, construct an original data set of the highway network, and compress the original data in the original data set of the highway network in blocks;
[0160] The data identification unit is used to encode the compressed regional highway network original data set using arithmetic logic coding to create data identification, and perform structured processing according to the identification.
[0161] The data adaptation and screening unit is used to screen the data in the regional highway network construction and maintenance data set according to the adaptability of the data, remove invalid data, and eliminate discrete and repeated data;
[0162] The road network maintenance data summary storage unit is used to summarize the regional highway network original data set after structured processing to obtain the regional highway network construction and maintenance data set.
[0163] like Fig. 9 As shown in the figure, the road network maintenance planning simulation module specifically includes:
[0164] A loss function construction unit, used to construct the discriminator loss function and the generator loss function respectively;
[0165] Training the discriminator unit, which is used to build a discriminator and train the discriminator based on the discriminator loss function;
[0166] The training generator unit is used to build a generator and train the generator based on the generator loss function;
[0167] A training and simulation model generation unit, which obtains a road network maintenance planning simulation model based on the trained discriminator and generator;
[0168] The road network maintenance planning simulation unit performs road network maintenance planning simulation based on the road network maintenance planning simulation model to obtain a simulation image set, and evaluates the maintenance plan based on the simulation image set and the road network maintenance planning data to obtain a road network maintenance planning simulation evaluation result.
[0169] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.
[0170] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A medium- and long-term planning decision-making method for highway network-level pavement maintenance, characterized in that: include: S1: Acquire original data of a regional highway network, pre-process the original data of the regional highway network, and store the data to obtain a regional highway network construction and maintenance data set; S2: Analyze and evaluate the regional highway network according to the regional highway network construction and maintenance data set to obtain a highway network evaluation result; S3: Obtaining road network maintenance planning data according to the regional highway network construction and maintenance data set and the highway network evaluation result; S4: performing simulation and simulation evaluation according to the road network maintenance planning data to obtain a road network maintenance planning simulation evaluation result; S5: Making a road network high-speed maintenance planning decision based on the road network maintenance planning simulation evaluation results.
2. A method for mid- and long-term planning and decision-making of highway network-level pavement maintenance according to claim 1, characterized in that: The S1 specifically includes: S101: Obtain the original data of the regional highway network and construct the original data set of the highway network; S102: compressing the original data in the highway network original data set into blocks; S103: using arithmetic logic coding to encode the compressed regional highway network original data set to create a data identifier, and performing structured processing according to the identifier, aggregating the structured regional highway network original data set to obtain a regional highway network construction and maintenance data set; S104: Screening the data in the regional highway network construction and maintenance data set according to the adaptability of the data, removing invalid data, and eliminating discrete and repeated data.
3. The method for medium- and long-term planning and decision-making of highway network-level pavement maintenance according to claim 2 is characterized in that: The S104 specifically includes: Adaptation detection is performed using the marked highway network data code to obtain a road network data entropy value table, and each entropy value with repetition and discreteness is listed according to the set entropy threshold MENT to screen the adaptation grouping function G, and the grouping function G is: In the formula, G() is the grouping function, x i is the i-th data column, i is the data column identifier, K n is the nth data set, K n+m is another data set, ENT(x i ) is the entropy value of the ith data column, MENT is the entropy threshold, and α is the threshold bias parameter.
4. The method for mid- and long-term planning and decision-making of highway network-level pavement maintenance according to claim 1 is characterized in that: The S3 specifically includes: S301: Constructing a road network maintenance planning data preprocessing analysis model; S302: Analyze the data in the regional expressway network construction and maintenance data set and the expressway network evaluation results according to the road network maintenance planning data preprocessing analysis model to obtain road network maintenance planning data.
5. A method for medium- and long-term planning and decision-making of highway network-level pavement maintenance according to claim 4, characterized in that: The S301 specifically includes: weight allocation, decision tree algorithm training weak differentiators, building loss function, updating weights, and weighted voting.
6. The method for mid- and long-term planning and decision-making of highway network-level pavement maintenance according to claim 1 is characterized in that: The S4 includes: Constructing a discriminator and a generator, and loss functions corresponding to the discriminator and the generator respectively; The discriminator and the generator are trained respectively using corresponding loss functions to obtain a trained discriminator and generator; Use the trained discriminator to assign probability values to the preset regional highway construction and maintenance planning simulation images; The trained generator enhances the data set of the regional highway construction and maintenance planning simulation image assigned with probability values according to the road network maintenance planning data to generate a simulation image; Acquire a set of simulation images according to the simulation images predicted by the generator; The maintenance plan is evaluated according to the simulation image set and the road network maintenance planning data to obtain a road network maintenance planning simulation evaluation result.
7. The method for mid- and long-term planning and decision-making of highway network-level pavement maintenance according to claim 1 is characterized in that: The S5 specifically includes: According to the simulation evaluation results of the road network maintenance planning, the road section properties, operating years and traffic volume of different road sections are obtained, and the maintenance type is determined according to the road section properties, operating years and traffic volume; Determine the influencing factors of different road sections according to the simulation evaluation results of the road network maintenance planning, and assign weights according to the importance, contribution and controllability of the influencing factors to obtain a maintenance demand index; According to the maintenance demand index combined with annual maintenance investment, maintenance countermeasure technology, maintenance management, and regional funds, the road network highway maintenance planning decision is obtained.
8. A medium- and long-term planning and decision-making system for highway network-level pavement maintenance, characterized in that: include: Data collection and storage management module, road network-level big data analysis and evaluation module, road network maintenance planning data preprocessing module, road network maintenance planning simulation module, road network high-speed maintenance planning decision module; The data collection and storage management module is used to obtain the original data of the regional highway network, pre-process the original data of the regional highway network, and store it to obtain a regional highway network construction and maintenance data set; A road network-level highway big data analysis and evaluation module is used to analyze and evaluate the regional highway network based on the regional highway network construction and maintenance data set to obtain a highway network evaluation result; A road network maintenance planning data preprocessing module, used to obtain road network maintenance planning data according to the regional highway network construction and maintenance data set and the highway network evaluation result; A road network maintenance planning simulation module is used to perform simulation and simulation evaluation based on the road network maintenance planning data to obtain a road network maintenance planning simulation evaluation result; The road network highway maintenance planning decision module is used to make road network highway maintenance planning decisions based on the road network maintenance planning simulation evaluation results.
9. The highway network-level pavement maintenance medium- and long-term planning decision system according to claim 8 is characterized in that: The data collection and storage management module specifically includes: A data block compression unit, used to obtain the original data of the regional highway network, construct an original data set of the highway network, and compress the original data in the original data set of the highway network in blocks; The data identification unit is used to encode the compressed regional highway network original data set using arithmetic logic coding to create data identification, and perform structured processing according to the identification. A data adaptation and screening unit, used to screen the data in the regional highway network construction and maintenance data set according to the adaptability of the data, remove invalid data, and eliminate discrete and repeated data; The road network maintenance data summary storage unit is used to summarize the regional highway network original data set after structured processing to obtain the regional highway network construction and maintenance data set.
10. The highway network-level pavement maintenance medium- and long-term planning decision system according to claim 8, characterized in that: The road network maintenance planning simulation module specifically includes: A loss function construction unit, used to construct the discriminator loss function and the generator loss function respectively; A training discriminator unit is used to construct a discriminator and train the discriminator based on the discriminator loss function; A training generator unit, used for constructing a generator and training the generator based on the generator loss function; A training and simulation model generation unit, which obtains a road network maintenance planning simulation model based on the trained discriminator and generator; The road network maintenance planning simulation unit performs road network maintenance planning simulation based on the road network maintenance planning simulation model to obtain a simulation image set, and evaluates the maintenance plan based on the simulation image set and the road network maintenance planning data to obtain a road network maintenance planning simulation evaluation result.
