Road traffic engineering test detection data acquisition and evaluation system

Abstract

The invention discloses a road traffic engineering test detection data acquisition and evaluation system, which relates to the technical field of traffic detection and comprises a data acquisition module, a data processing and analysis module, an evaluation index matching module, a data evaluation module and a result output module. According to the invention, automatic identification and accurate evaluation can be carried out according to characteristics of different types of roads, acceptance of newly built roads and reconstruction of old roads are ensured to reach the standard, and the limitation of traditional unified standards is overcome; multi-dimensional data acquisition is comprehensive and highly targeted, and evaluation deviation is avoided; indexes are comprehensive and can be dynamically adjusted, the reality of new and old roads is met, and evaluation effectiveness is improved; and single and comprehensive evaluation optimization is realized, multiple factors are considered, and a comprehensive and objective basis is provided for road acceptance and reconstruction.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic detection, and particularly to a highway traffic engineering test detection data acquisition and evaluation system. Background Art

[0002] In the field of highway traffic engineering, the test detection and evaluation of highways are crucial, which are related to many aspects such as the quality, safety and subsequent use efficiency of highways. However, there are many limitations in the traditional test detection and evaluation methods of highway traffic engineering, which are difficult to meet the actual needs of diversified highway construction and renovation projects.

[0003] I. Lack of targeted evaluation ability

[0004] Traditional traffic acceptance detection methods often focus on a specific type of highway, or it is difficult to be effectively applied to other different types of highways when applicable to one type of highway. For example, in terms of administrative level division, national highways, as the main trunk roads with national political and economic significance, have significant differences in construction standards, functional requirements from county roads, township roads, village roads, etc. County roads mainly serve the traffic connection within the county, township roads focus on the connection between villages and with the outside, and village roads directly serve rural production and life. Their respective importance, traffic flow characteristics and impacts on the surrounding environment are all different. However, traditional detection methods usually adopt unified standards and processes, and cannot conduct targeted evaluations according to the characteristics of highways at different administrative levels, making it difficult to accurately judge whether the newly built national highway meets its high-standard acceptance requirements, and whether the county roads, township roads, etc. after the renovation of old roads meet the corresponding renovation standards.

[0005] Similarly, in terms of technical level division, expressways, first-class highways, second-class highways, third-class highways and fourth-class highways have their own characteristics in terms of traffic volume adaptation, design speed, lane setting and requirements for driving safety and traffic capacity. For example, expressways need to meet the conditions of high speed, large traffic capacity and full closure and full interchange, while fourth-class highways, as the branch roads connecting counties, townships, villages, etc., have relatively small traffic flow and low design speed. Traditional detection means fail to fully consider these technical level differences, and when evaluating newly built or renovated highways at different technical levels, they cannot accurately judge based on their respective key indicators, and cannot accurately know whether the newly built expressway meets its strict acceptance specifications, and whether the third-class highways, etc. after the renovation of old roads reach the renovation goals.

[0006] From the perspective of function and environment, various types of roads such as ordinary roads, urban roads, mountain roads, coastal roads, forest roads, elevated roads, and tunnel roads also have their unique construction requirements and usage characteristics. Urban roads need to focus on meeting the complex traffic needs of pedestrians and vehicles within the city. Mountain roads should fully consider the construction difficulties and safety guarantees brought by the terrain and landform. Coastal roads have to deal with special factors such as marine climate. However, traditional detection methods lack careful consideration of these functional and environmental differences, cannot conduct personalized evaluations for roads of different functional and environmental types, and are unable to effectively determine whether a newly built mountain road meets the construction standards under complex terrains, and whether the coastal road after old road reconstruction meets the operation requirements after reconstruction.

[0007] II. Single Data Collection and Evaluation Criteria

[0008] Traditional detection methods are relatively single in data collection. They can often only obtain limited types of data and cannot comprehensively cover all kinds of key information required during the construction of new roads and the reconstruction of old roads. For example, for newly built roads, they may only focus on some basic physical index data such as road surface flatness, while ignoring data collection on aspects such as the impact of construction noise and dust on surrounding residents and the ecological environment during the construction process. For old road reconstruction, they may simply understand the road surface disease conditions but do not deeply collect data related to traffic guidance capabilities, such as temporary traffic control time and traffic congestion index. This makes the collected data unable to fully reflect the actual situation of the road and makes it difficult to accurately judge whether the newly built road can pass the acceptance detection standards and whether the old road reconstruction meets the standards based on this.

[0009] Moreover, traditional detection methods are also relatively fixed in evaluation criteria. Usually, a unified set of indicators is used to measure all roads without setting different evaluation indicators according to the characteristics of different roads. Whether it is a newly built road or an old road reconstruction, different types of roads (such as classified by the above various methods) have their own different key parameters and potential problems during construction and use, and corresponding evaluation indicators are needed for accurate evaluation. However, traditional methods cannot achieve this, resulting in evaluation results that often cannot truly reflect the actual quality and compliance of the road and cannot meet the needs of accurate evaluation of different roads.

[0010] III. Insufficient Automated Evaluation Capability

[0011] Existing detection methods lack the ability to automatically evaluate roads according to different evaluation criteria based on the collected data. When faced with new road construction and old road renovation projects, it is impossible to automatically identify the road type based on the collected data and automatically carry out the evaluation work according to the corresponding different indicators. For example, when the detection data involves multiple road types (such as both new national road data and rural road data for old road renovation), the traditional method cannot automatically distinguish and adopt appropriate evaluation criteria for separate evaluation, but relies on manual classification and judgment. This not only has low efficiency but also is prone to human errors, unable to ensure the accuracy and objectivity of the evaluation results, and difficult to meet the growing demand for efficient and accurate evaluation in road construction and renovation projects.

[0012] In summary, there are many limitations in the traditional road traffic engineering test detection and evaluation methods, which cannot meet the needs of new road construction and old road renovation projects in terms of targeted evaluation, data collection and evaluation criteria, and automated evaluation. Therefore, it is necessary to design a "road traffic engineering test detection data collection and evaluation system" applicable to new road construction and old road renovation. This system can accurately evaluate the corresponding roads (including roads divided by administrative level, technical level, function, and environment) according to different evaluation criteria based on the collected data, so as to solve the problems existing in the prior art and ensure the quality and efficiency of road construction and renovation projects.

[0013] In view of this, a road traffic engineering test detection data collection and evaluation system is provided to overcome the above problems. Summary of the Invention

[0014] The purpose of the present invention is to provide a road traffic engineering test detection data collection and evaluation system to solve the problems raised in the above background technology.

[0015] To solve the above technical problems, a road traffic engineering test detection data collection and evaluation system provided by the present invention includes a data collection module, a data processing and analysis module, an evaluation index matching module, a data evaluation module, and a result output module:

[0016] The data collection module includes an on-vehicle sensor unit. The on-vehicle sensor unit includes a material composition detection sensor for detecting the composition ratio of newly paved road surface materials and a pavement structure layer detection radar for detecting the thickness change and internal disease conditions of the old road structure layer; the data processing algorithm of the on-vehicle data collection terminal analyzes the deviation between the newly built road material composition data and the preset design value in real time and gives adjustment suggestions, and uses the old road structure layer detection radar data combined with machine learning algorithms to identify the disease type and degree.

[0017] The data processing and analysis module includes a data cleaning and fusion unit and a feature extraction and highway type identification unit. For newly built highway data, a method based on the construction process model is used to identify abnormal data caused by construction process fluctuations. For data on old road reconstruction, historical disease data and current environmental data are combined to identify abnormal data caused by environmental changes. Feature extraction and highway type identification unit: For newly built highways, construction process features are extracted, including but not limited to the vibration frequency features of construction machinery and the transportation path features of construction materials. For old road reconstruction, features related to the disease history are extracted, including but not limited to the time of first appearance of the disease and the disease development speed.

[0018] The evaluation index matching module includes an evaluation index library management unit and an automatic matching unit. The evaluation index library management unit: For newly built highways, the evaluation index library includes indicators of the impact of construction noise and dust on surrounding residents and the ecological environment. For old road reconstruction, it includes traffic guidance ability evaluation indicators, including but not limited to temporary traffic control time and traffic congestion index. The automatic matching unit, when matching evaluation indexes and weights, considers construction progress and expected service life factors for newly built highways, and combines historical evaluation data of old roads and the priority of current reconstruction goals for old road reconstruction.

[0019] The data evaluation module includes a single-index evaluation unit and a comprehensive evaluation unit. The single-index evaluation unit: For complex indexes of newly built highways, an evaluation method combining a dedicated laboratory simulation model and on-site collected data is established. For disease repair indexes in old road reconstruction, non-destructive testing technology is introduced. The comprehensive evaluation unit considers social impact factors in the analytic hierarchy process for newly built highways and old road reconstruction, and optimizes the fuzzy set partitioning method in fuzzy comprehensive evaluation; analyzes the sensitivity of each evaluation index to the comprehensive result.

[0020] The result output module includes a visualization display unit and a report generation unit.

[0021] Furthermore, the data acquisition module also includes a drone remote sensing unit: The drone is equipped with an optical camera with infrared thermal imaging function and a LiDAR device with micro displacement monitoring ability, and is equipped with a professional drone takeoff and landing platform and a control device that can achieve automatic flight control; and an algorithm for planning flight routes according to the key areas of newly built highway construction and newly paved road surfaces, the distribution of old road diseases and the key areas of reconstruction is introduced, which is used to regularly inspect the highway and transmit the collected image data and point cloud data back to the receiving device of the acquisition vehicle in real time.

[0022] Furthermore, the data acquisition module also includes a satellite remote sensing data receiving unit: It includes a receiving antenna and decoding device for receiving synthetic aperture radar data, and data preprocessing software; the data preprocessing software automatically identifies and removes the influence of atmospheric interference and terrain shadows in satellite remote sensing data using artificial intelligence algorithms.

[0023] Further, the vehicle-mounted sensors are connected to the vehicle-mounted data acquisition terminal through the in-vehicle wiring system. The vehicle-mounted data acquisition terminal includes a data preprocessing subunit and a data compression subunit.

[0024] Further, in the data cleaning and fusion unit, a multi-source data fusion algorithm based on deep learning is adopted to control the fusion accuracy of data from different sources in space and time, assign high-level weights to the real-time data during the construction of new roads, and weight the long-term historical data during the renovation of old roads.

[0025] Further, for the feature extraction and road type recognition unit, an adversarial learning mechanism for improving the recognition ability of complex features in the construction of new roads and the renovation of old roads is introduced during the training of the machine learning model.

[0026] Further, an index matching assistance mechanism based on risk assessment is introduced in the automatic matching unit to dynamically adjust the indexes and weights according to the environment where the road is located and the abnormal conditions during the data acquisition process.

[0027] Further, an index evaluation subunit for analyzing the changing trend of the data indexes of new roads and the renovation of old roads over time is introduced in the single-index evaluation unit.

[0028] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0029] It can automatically identify and accurately evaluate according to the characteristics of different types of roads (classified in multiple ways), ensure the acceptance of new roads and the compliance of the renovation of old roads, and overcome the limitations of the traditional unified standard.

[0030] Collect data in multiple dimensions, which is comprehensive and highly targeted, avoiding evaluation deviation.

[0031] The indexes are comprehensive and can be dynamically adjusted, meeting the actual conditions of new roads and old roads, and improving the effectiveness of evaluation.

[0032] The single-index and comprehensive evaluations are optimized, considering multiple factors, providing a comprehensive and objective basis for the acceptance and renovation of roads. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of a data acquisition and evaluation system for highway traffic engineering test and detection according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0035] Please refer to Figure 1 , the present invention provides a technical solution:

[0036] Refer to Figure 1 , an embodiment of a highway traffic engineering test and detection data acquisition and evaluation system:

[0037] Data acquisition module

[0038] Vehicle-mounted sensor unit

[0039] Material composition detection sensor (for new roads):

[0040] Let the designed proportion of a certain key component (such as a certain additive in asphalt) in the newly paved road surface material be P d , during the construction process, the material composition detection sensor collects the actual proportion value P of this component every fixed time interval Δt (for example, Δt = 5 minutes) i (t) (t represents the collection time). The formula for calculating the deviation value is:

[0041]

[0042] When E(t) exceeds the preset threshold (such as 0.03P d ), it is determined that there may be a problem with the material ratio. According to the empirical formula established based on a large amount of past data, the adjustment suggestion can be expressed as: A = f(E(t), P d , t), where f is a function fitted based on historical construction data and is used to guide operations such as adjusting the material addition amount.

[0043] Timely discover problems with the material ratio of new roads and ensure the construction quality. Through real-time monitoring and adjustment, avoid quality problems caused by unqualified materials. The long-term accumulated P i (t) data, combined with environmental data such as temperature H and humidity in different regions (obtained through other sensors or external data sources), can analyze the change law of material performance. For example, it is found that the relationship between the material performance change rate R and environmental factors and composition ratio changes is: R = g(P i (t), H, t), providing a basis for material research and development.

[0044] Road surface structure layer detection radar (for old road reconstruction):

[0045] Let the initial thickness of the old road structure layer at a certain position be \(T_0\), and the thickness values detected by the detection radar at different times \(t\) be \(T(t)\). The disease degree index \(D\) can be calculated by the following formula:

[0046]

[0047] where \(t_1\) and \(t_2\) are the detection time periods, and \(w(t)\) is a function considering time weights, reflecting the time characteristics of disease development. Using machine learning algorithms (such as support vector machines), \(D\) and other relevant data (such as radar reflection wave characteristics, etc.) are used as inputs to identify the disease type.

[0048] Precisely locate the deep diseases of the old road, provide an accurate basis for the renovation plan, and improve the renovation effect. Combining the traffic flow data \(F(t)\) (obtained by other means), analyze the relationship between the disease development trend and traffic flow. For example, the disease development speed \(V\) D The relationship with traffic flow is:

[0049] V D = h(F(t), D), providing a reference for preventive maintenance.

[0050] UAV remote sensing unit

[0051] Optical camera (infrared thermal imaging function):

[0052] In a newly built highway, let the ideal temperature uniformity index after the pavement material is laid be \(U\) id , and the temperature values at different positions obtained by the infrared thermal imaging camera be \(T\) ij (\(i\) represents the row coordinate, \(j\) represents the column coordinate). Calculate the temperature uniformity index \(U\):

[0053]

[0054] where \(n\) is the number of measurement points, is the average temperature. If \(U > U\) id , it indicates that the material laying temperature is uneven. In the renovation of the old road, similarly, different temperature threshold ranges can be set for the disease area and the normal area, and the disease situation can be judged by comparing the detected temperatures.

[0055] LiDAR device (micro displacement monitoring):

[0056] For newly built highway structures (such as bridges), let the initial position coordinates of a certain point on the structure be \((x_0, y_0, z_0)\), and the coordinates at different times \(t\) during the construction process be \((x(t), y(t), z(t))\). The formula for the micro displacement amount \(S(t)\) is:

[0057]

[0058] When it exceeds the preset threshold (determined according to the design requirements of the structure), there may be potential structural safety hazards. In the renovation of old roads, it can be used to monitor the position changes after structural reinforcement and evaluate the reinforcement effect.

[0059] UAV flight route planning algorithm:

[0060] For newly built roads, according to the construction progress plan and the coordinate information of key construction areas (such as the positions of bridges and tunnels), plan the flight route R nem , so that the UAV can preferentially cover these areas. For the renovation of old roads, combine the old road disease distribution map (drawn through the previous detection data) and the coordinates of the key renovation areas (such as severely diseased sections) to plan the route R old . The flight route planning needs to meet the optimization goal of covering all key areas and having the shortest flight time, which can be solved by genetic algorithms and other methods.

[0061] Timely detect material and structural problems in newly built roads to ensure construction safety and quality; accurately evaluate the reinforcement effect and disease repair situation in the renovation of old roads. The long-term accumulated infrared thermal imaging temperature data can be used to study the influence of highway temperature changes on pavement performance under different seasons (season factor S) and weather conditions (weather factor W), such as the relationship between the pavement performance change index P and temperature, season, and weather:

[0062] R = k(U, S, W).

[0063] The tiny displacement data monitored by the LiDAR device can be used as the basis for long-term structural health monitoring and for formulating highway maintenance cycles. By analyzing the trend T r (S) of displacement changes over time, determine the maintenance cycle C:

[0064] C = m(T r (S)).

[0065] Satellite remote sensing data receiving unit

[0066] Synthetic aperture radar (SAR) data reception and processing:

[0067] In the construction area of newly built roads, set the allowable threshold of land settlement amount as. Obtain the land height values at different times through SAR data, and calculate the settlement amount L(t) = H(t0) - H(t) (t0 is the initial time). If L(t) > L th , it may have an impact on the construction and early warning is required. In the renovation of old roads, similarly analyze the influence of the surrounding environment changes on the development of old road diseases. For the influence of atmospheric interference and terrain shadows, use artificial intelligence algorithms (such as neural networks) to establish a removal model. Let the original satellite remote sensing data be D raw , and the processed interference-free data be D clean , and the neural network model N satisfies: Dcleam = N(D raw ).

[0068] In newly built roads, it can early warn of construction problems caused by changes in the surrounding environment in advance to ensure the smooth progress of construction; in the renovation of old roads, it can more accurately analyze the relationship between diseases and the environment and optimize the renovation plan. Long-term SAR data can be used to study the long-term change trend of the geological environment around the road. For example, the relationship between geological change indicators and data such as land subsidence and water source change is: (for water source change data), which provides a macro reference for highway planning and design. The improved data preprocessing algorithm can be applied to the processing of satellite remote sensing data in other fields to improve data availability.

[0069] Data Processing and Analysis Module

[0070] Data Cleaning and Fusion Unit

[0071] Data Cleaning (Newly Built Road):

[0072] According to the construction process model, the normal data range under a certain construction stage S is set as For the collected data D, if or then it is determined as abnormal data. For example, in the concrete pouring stage, the concrete strength data should be within a certain range, and if it exceeds, it is abnormal. For the data of old road renovation, let the historical disease data feature vector be H = (h1, h2,... h n ), the current environmental data vector be E = (e1, e2,... e m ), and a logistic regression model is established to judge whether the data is abnormal. Let the abnormal probability be: P abnormal , then:[[]]

[0073]

[0074] (a i , b j , c are model parameters), when P abnormal exceeds a certain threshold (such as 0.5), the data is determined as abnormal.

[0075] Data Fusion (Based on Deep Learning):

[0076] Let the in-vehicle sensor data be X = (x1, x2,..., x p ), the UAV remote sensing data be Y = (y1, y2,..., y q ), and the satellite remote sensing data be Z = (z1, z2,..., z r ). Using a deep learning fusion algorithm (such as a convolutional neural network), the data is fused into a feature vector F. The loss function L of the network is designed as:

[0077]

[0078] where N is the number of samples, represents the weighted fusion operation, w x , w y , w z , are weights, and λ is the regularization parameter. During the fusion process, weights w new are assigned to the real-time data in the construction of new roads, old and weights w

[0079] Improve the accuracy of data cleaning, reduce misjudgments, and ensure the quality of subsequent analysis data; more accurate data fusion provides a reliable basis for highway type identification and evaluation, especially for highways in complex environments. The model-based data cleaning method can be fed back to the construction process to optimize the construction technology. For example, adjust the construction operation parameters according to the frequency of abnormal data. The advanced data fusion algorithm can be extended to other multi-source data fusion scenarios, such as urban comprehensive monitoring systems. By adjusting the network structure and parameters, it can be applied to data fusion problems in different fields.

[0080] Feature extraction and highway type identification unit

[0081] Feature extraction (new road):

[0082] Let the vibration frequency characteristics of construction machinery be V = (v1, v2,..., v m ), and the transportation path characteristics of construction materials be P = (p1, p2,..., p n ), (transportation path information is obtained through positioning technologies such as GPS). For highway type identification, these characteristics are combined with other data (such as road design parameters, etc.) to form a feature vector T new = (V, P,...). For the renovation of existing roads, let the time of the first appearance of diseases be t first , and the disease development speed be v disease , to form a feature vector T old = (t first , v disease ,...).

[0083] Highway type identification (machine learning model based on adversarial learning mechanism):

[0084] Let the generator network be G and the discriminator network be D. For the feature data T of new road construction and existing road renovation, the generator generates fake data T fake = G(z), (z is a random noise vector), and the discriminator determines whether the input data is real or fake. The goal is to make the discriminator unable to distinguish between real and fake data. At the same time, minimize the cross-entropy loss between the feature data and the real highway type label y;

[0085] To train the model. During the training process, the model parameters are continuously adjusted to improve the recognition ability of complex features, especially the recognition of highway types under different geographical environments and working conditions.

[0086] More comprehensive feature extraction helps to accurately identify highway types, especially for new construction and old road renovation situations, providing accurate inputs for subsequent evaluation index matching. Construction process features can be used to evaluate construction efficiency and quality control levels. For example, the relationship between the vibration frequency of construction machinery and the construction quality index Q: Q = r(V). The model trained by the adversarial learning mechanism can be used for other similar classification problems, such as the recognition of different types of building structures, and different domain classification tasks can be achieved by adjusting features and network structures.

[0087] Evaluation Index Matching Module

[0088] Evaluation Index Library Management Unit

[0089] Index Addition and Dynamic Adjustment:

[0090] For newly built highways, let the construction noise index be N, the dust emission index be D, the degree of impact on surrounding residents be I, and the ecological environment impact index be E. These index values are calculated through sensors and environmental assessment models. The degree of impact of construction noise on surrounding residents can be calculated by the formula:

[0091] Calculated (d(t) is the resident distribution density function). For old road renovation, let the temporary traffic control time be T ctrl , and the traffic congestion index be C jarn , obtained through traffic monitoring data. The evaluation index library is adjusted in real time according to changes in policies and regulations in different regions and the latest scientific research results. Let the adjustment function be U, and the content of the index library be I lib The update formula at time t is:

[0092] I lib (t) = U(I lib (t - 1), R(t), S(t)).

[0093] (R(t) is the information on changes in policies and regulations, S(t) is the information on scientific research results). At the same time, users feedback on the rationality of the indexes through the online platform. Let the feedback impact factor be, then the adjustment formula can be further modified as:

[0094] I lib (t) = U(I lib (t - 1), R(t), S(t), F(t)).

[0095] Make the evaluation indicators more comprehensive, meet the actual needs of new road construction and old road reconstruction, and consider important factors such as environment and traffic; the dynamic adjustment mechanism ensures the timeliness and scientificity of the indicator library. The data of environmental impact evaluation indicators can be used to evaluate the long-term impact of highway construction on the regional ecology. For example, the long-term change index E of the ecological environment long Cumulative relationship with ecological indicators during the construction process: The user feedback mechanism can promote the continuous optimization of the indicator library, form a consensus on evaluation standards within the industry, and improve the rationality of indicators and weight settings by analyzing a large amount of user feedback data.

[0096] Automatic matching unit

[0097] Indicator and weight matching (newly built highway):

[0098] When matching evaluation indicators and weights, for newly built highways, set the construction progress factor as P prog , and the expected service life factor as L exp . The indicator weight W new is determined according to these two factors and other relevant factors (such as highway grade, etc.). For the weight of the pavement evenness indicator It is calculated through the formula:

[0099]

[0100] (G is the highway grade). For old road reconstruction, set the historical evaluation data vector of the old road as H = (h1, h2,..., h n ), and the current reconstruction target priority vector as P = (p1, p2,..., p m ). If the main goal of old road reconstruction is to improve the bearing capacity, the weight of indicators related to structural strength is correspondingly increased. Set the weight of the structural strength indicator Its calculation can be associated with the goal of improving the bearing capacity: (C is the bearing capacity indicator). At the same time, based on the risk assessment-based indicator matching auxiliary mechanism, set the environmental risk factor of the highway as R env (such as natural disaster risk, traffic volume), and the data collection anomaly factor as A data . The formula for dynamically adjusting the indicator weight W is:

[0101] W = W0 + k1R env + k2A data ;

[0102] (W0 is the initial weight, k1, k2 are adjustment coefficients).

[0103] More precisely match the evaluation indicators and weights to make the evaluation results conform to the actual situation and improve the evaluation effectiveness. The matching mechanism based on risk assessment can help detect potential problems in advance and provide a basis for highway risk management. For example, formulate countermeasures in advance according to the risk assessment results to reduce disaster losses. The matching method considering the construction progress and the priority of renovation goals can be extended to other engineering project evaluation systems and can be applied to similar project evaluations by adjusting relevant factors and formulas.

[0104] Data evaluation module

[0105] Single - index evaluation unit

[0106] Evaluation of complex indicators for newly built highways (combining laboratory simulation and field data):

[0107] For the performance evaluation of new materials for newly built highways, let the performance index obtained from the laboratory simulation model be P lab , and the performance index obtained from the field - collected data be P field . The comprehensive evaluation index is obtained through weighted fusion:

[0108] P new = w1P lab + w2P field ;

[0109] (w1, w2 are weights, determined according to the reliability of simulation and field data). For the disease repair index in the renovation of old roads, use non - destructive testing techniques (such as ground - penetrating radar), and let the repair layer quality index be Q repair , and calculate according to data such as the characteristics of radar reflection waves. For example, by establishing a relationship model between the reflection wave intensity and the repair quality, calculate Q repair . At the same time, during the single - index evaluation process, analyze the trend of index data over time. Let the index data of the newly built highway be D new (t), and the index data of the old - road renovation be D old (t). By fitting the time - series data, obtain the trend functions Tr new (t) and Tr old (t), such as using linear regression or polynomial fitting methods.

[0110] Evaluate single - indicators more accurately, especially complex indicators and the disease repair situation of old roads; data trend analysis helps to detect problems in time and take preventive measures. The method of combining laboratory simulation models and field data can be used for the performance evaluation of new materials in other application fields. By adjusting model parameters and weights, it can be applicable to different materials and application scenarios. The application experience of non - destructive testing techniques in the renovation of old roads can be extended to the repair evaluation of other infrastructures. For example, similar non - destructive testing techniques and evaluation methods can be applied in bridge repair.

[0111] Comprehensive Evaluation Unit

[0112] Analytic Hierarchy Process and Consideration of Social Impact Factors (New Road Construction and Existing Road Reconstruction):

[0113] In the Analytic Hierarchy Process, for new road construction and existing road reconstruction, assume that the social impact factors include the promotion index E for local economic development eco , and the impact index C on the convenience of residents' travel travel . Construct a hierarchical structure model and layer these social impact factors together with other evaluation indicators. Assume that the goal layer is the highway quality evaluation Q, the criterion layer includes different types of evaluation index categories (such as pavement condition, structural safety, environmental impact, social impact, etc.), and the scheme layer is specific evaluation indicators (such as pavement smoothness, structural strength, construction noise, E eco , C travel , etc.). Determine the relative importance between elements of each layer through the pairwise comparison method and construct a judgment matrix A. For example, when comparing the social impact factors with other criterion layer factors, if the importance of the social impact factors relative to the pavement condition is considered to be 3 (determined based on expert opinions or actual investigations), then the corresponding element a ij = 3 (i represents the row where the social impact factors are located, and j represents the column where the pavement condition is located). By solving the maximum eigenvalue and eigenvector of the judgment matrix, obtain the weight vector of each index. In the fuzzy comprehensive evaluation, optimize the partitioning method of the fuzzy set. Assume that the evaluation grade fuzzy set is:

[0114] V = {v1, v2, v3, v4} (such as excellent, good, qualified, unqualified). For each evaluation index, determine its membership function for each evaluation grade according to its actual value and evaluation criteria. For example, for the pavement smoothness index, assume its actual measured value is x, and the membership function can be determined according to the threshold range of smoothness. For example, when within the excellent threshold range, and other membership degrees are 0. Establish a fuzzy relation matrix R, where the element r ij represents the membership degree of the i-th index to the j-th evaluation grade. Through the fuzzy composition operation (here different composition operators can be used, such as the dominant factor prominent type operator or the weighted average type operator) to obtain the comprehensive evaluation result vector B, and its elements represent the degree to which the highway quality belongs to each evaluation grade. Establish a sensitivity analysis function for the comprehensive evaluation result and calculate the sensitivity of each evaluation index to the comprehensive result. Assume that the sensitivity coefficient S i of the comprehensive evaluation result B to the index i can be determined by calculating the change ΔB in the comprehensive evaluation result caused by the change Δw i in the weight of the index i. For example, (finite difference and other methods can be used for approximate calculation).

[0115] Make the comprehensive evaluation more comprehensive, consider social impact factors, and conform to the practical significance of highway construction and renovation; sensitivity analysis helps to optimize the evaluation index system and improve the scientificity of the evaluation. The evaluation method of social impact factors can be used for the evaluation of other infrastructure projects. For example, in the construction of urban rail transit, the impact on the economic development along the line and the travel of residents can be considered similarly. The results of sensitivity analysis can guide the resource allocation and key attention directions in the process of highway construction and renovation. For example, if the sensitivity coefficient of a certain index is relatively high, it indicates that it has a greater impact on the comprehensive evaluation result, and the construction or renovation work related to this index can be preferentially guaranteed under limited resources.

[0116] Result output module

[0117] Visual display unit

[0118] Animation display of construction technology (for new highways) and animation display of the historical development of diseases (for old road renovation):

[0119] In the interactive visualization interface, for new highways, through the animation display function of construction technology, the construction process is decomposed into multiple key steps S1, S2, …, S n . Each step is displayed in the form of an animation, including details such as the operation of construction machinery and the laying of materials. For example, in the step of bridge erection, the lifting action of the crane and the splicing process of steel girders are shown. For old road renovation, according to the historical data of disease development, the generation and development process of diseases are shown in an animation in time series. Let the disease development stages be D1, D2, …, D m , and each stage is represented by different colors or marks in the animation. For example, the initial cracks are represented by red thin lines, and as time goes by, the cracks expand and are represented by thicker red lines. Using augmented reality (AR) technology, when viewing the highway on site, relevant evaluation results and data information are displayed in real time through a mobile device. Let the current position coordinates obtained by the mobile device be (x m , y m ). By matching with the coordinates of the evaluation data, the evaluation results of the surrounding area are displayed on the device screen. For example, when viewing the construction site of a new highway, the index evaluation situation of the current construction area is displayed, such as the deviation value of material composition and the micro displacement of the structure; when viewing the renovated section of an old road, the disease repair situation (such as the quality index of the repair layer) and the remaining problems (such as the location and severity of the unrepaired diseases) are shown.

[0120] Enhance the visualization effect to enable users to more intuitively understand the construction process of the newly built highway and the evolution of old road diseases, facilitating analysis and decision-making; the AR technology improves the information richness of on-site viewing and enables staff to obtain data in a timely manner. The construction process animation can be used for the training of construction workers, and new employees can quickly understand the construction process and key technical links by watching the animation. The animation of the disease development history can be used as educational materials for studying the formation mechanism of diseases, and researchers can summarize the disease development rules by analyzing the animation. The application experience of AR technology in the highway field can be extended to on-site monitoring in other engineering fields, such as construction sites of buildings and water conservancy projects, and different engineering scenarios can be applied by adjusting the display content and data matching methods.

[0121] Report generation unit

[0122] Content addition and online sharing and collaborative editing functions:

[0123] Add content on quality control measures and effect evaluation during the construction process of the newly built highway, as well as traffic organization plans and implementation effect analysis during the old road reconstruction process to the report. For the quality control measures of the newly built highway, details of quality inspection points, inspection methods and frequencies during the construction process shall be recorded. For example, during the concrete pouring process, the number of times of sampling and testing the strength per cubic meter of concrete, the standard testing method, etc. Let the implementation effect index of the quality control measures be E qc , which is calculated by comparing the quality inspection results with the design requirements. For the traffic organization plan of the old road reconstruction, information such as the time, scope and diversion measures of traffic control shall be recorded. Let the implementation effect index of the traffic organization plan be E to , which is evaluated through traffic flow data, congestion index changes, etc. At the same time, the selection process of evaluation indicators and the determination process of weights shall be explained in detail, including the meaning, source of each indicator and the method of determining weights (such as the pairwise comparison process in the analytic hierarchy process). Implement the online sharing and collaborative editing functions of the report. Through the network platform, personnel from different departments and units (such as the construction party, supervision party, design party, management department, etc.) can access and edit the report simultaneously. Let users be U1, U2, …, U k , and each user's editing operations on the report be O1, O2, …, O m (such as adding content, modifying data, auditing, etc.), and the legality of operations and data consistency shall be ensured through the permission management system. For example, the construction party can add actual data and situation descriptions during the construction process, the supervision party can audit the data and put forward opinions, and the design party can give optimization suggestions for the design scheme based on the evaluation results, etc.

[0124] The report content is more comprehensive, providing more detailed basis for subsequent decision-making in highway construction and renovation; the online sharing and collaborative editing functions improve work efficiency and promote multi-department cooperation. The quality control measures and the evaluation content of the traffic organization plan can be used as reference cases for similar projects, and other highway construction or renovation projects can draw on these contents to optimize their own quality control and traffic organization work. The collaborative editing mode of the report can be extended to other document preparation scenarios that require multi-department cooperation, such as large-scale construction projects, urban comprehensive development projects, etc., to achieve efficient document collaborative preparation by establishing a similar network platform and permission management system.

[0125] Summary:

[0126] A variety of sensors and data collection methods are set respectively for newly built highways and old road renovations. For newly built highways, material composition detection sensors are used to monitor the material ratio in real time (such as the proportion of asphalt additives), optical cameras (infrared thermal imaging) are used to check the uniformity of the paving temperature of road surface materials, LiDAR equipment is used to monitor the tiny displacement of structures, and satellite remote sensing is used to monitor the land settlement in the construction area. In old road renovations, ground-penetrating radar for road surface structure layers is used to detect the thickness changes of the structure layers. Combining traffic flow data, disease information and surrounding environment data are comprehensively obtained through drone remote sensing and satellite remote sensing technologies. Different types of highways (newly built and old roads) and highways with different grades and functions have different key parameters and potential problems during construction and use. The traditional unified data collection method cannot meet the special needs of newly built highways and old road renovations respectively, and targeted collection means are required to obtain accurate information and provide reliable data for subsequent evaluations. This multi-dimensional and targeted data collection method can comprehensively and accurately obtain various data during the construction of newly built highways and old road renovations, effectively avoiding evaluation deviations caused by missing or inaccurate data, thus providing strong support for judging whether newly built highways meet the acceptance standards and whether old road renovations are up to standard.

[0127] In terms of data cleaning, newly built highways judge abnormal data based on the construction process model, and old road renovations make judgments through a logistic regression model combined with historical diseases and current environment data. Data fusion uses deep learning algorithms to assign different weights to the real-time data of newly built highway construction and the long-term historical data of old road renovations according to the timeliness and importance of the data. At the same time, by extracting features such as the vibration frequency of construction machinery, the material transportation path, the occurrence time and development speed of diseases, a machine learning model based on the adversarial learning mechanism is used to accurately identify the highway type. The data quality is improved, misjudgments are reduced, and a reliable basis is provided for highway type identification and evaluation, especially for highways in complex environments. It can more accurately reflect the actual situation of newly built highways and old road renovations, making the evaluation results closer to the real state and helping to accurately judge whether the highways meet the standards.

[0128] The evaluation index library management unit adds indicators such as construction noise, dust, the degree of impact on surrounding residents, and ecological environment impact for newly built roads, and calculates their values through sensors and environmental assessment models. For the renovation of existing roads, indicators such as temporary traffic control time and traffic congestion index are included, and the content of the index library is adjusted in real time according to changes in local policies and regulations, the latest scientific research results, and user online feedback. In the automatic matching unit, the index weights for newly built roads are determined according to factors such as construction progress, expected service life, and road grade, and for the renovation of existing roads, the weights are dynamically adjusted according to the historical evaluation data of the existing roads, the priority of renovation goals, and environmental risks and data collection anomalies. There are significant differences between newly built roads and the renovation of existing roads in terms of the construction process and their impact on the surrounding environment. Traditional traffic acceptance inspection indicators cannot fully cover these special situations. It is necessary to add and adjust evaluation indicators according to their respective characteristics and reasonably determine the weights to make the evaluation results conform to the actual situation. Make the evaluation indicators more comprehensive, scientific, and reasonable, meet the actual needs of newly built roads and the renovation of existing roads, can accurately reflect various situations in the process of road construction and renovation, including the impact on the environment and traffic, improve the effectiveness of the evaluation, and ensure that the evaluation results can truly reflect whether the road meets the corresponding standards.

[0129] In the evaluation of individual indicators, for newly built roads, the performance evaluation of new materials combines laboratory simulation and on-site collected data, and for the renovation of existing roads, non-destructive testing technology is used to evaluate the disease repair indicators, and the trend of index data changing over time is analyzed. In the comprehensive evaluation, social impact factors (such as the promotion of local economic development and the convenience of residents' travel) are considered, a hierarchical structure model is constructed through the analytic hierarchy process to determine the weights of each indicator, the fuzzy comprehensive evaluation method is used to optimize the evaluation process, and sensitivity analysis is carried out. The use of new materials in newly built roads and the disease repair of existing road renovations require special evaluation methods to accurately assess their quality. At the same time, road construction and renovation should not only consider their own physical indicators, but also consider the impact on society, and comprehensive evaluation is required to comprehensively measure the value and compliance of the road. Sensitivity analysis helps to optimize the index system. Evaluate individual indicators more accurately, especially complex indicators and the disease repair situation of existing roads. Considering various factors comprehensively makes the evaluation more comprehensive and scientific, and can provide a more comprehensive and objective basis for the acceptance of newly built roads and the compliance of existing road renovations, guiding resource allocation and the direction of key attention.

[0130] The visualization display unit shows the construction process of the newly built road through construction technology animations and the disease evolution history of the old road through disease development history animations, and uses augmented reality (AR) technology to display the evaluation results and data information in real time when viewing the road on site. The report generation unit adds the content of quality control measures and effects of the newly built road, traffic organization plans and implementation effects of the old road reconstruction, and details the process of selecting evaluation indicators and determining weights. At the same time, it realizes the online sharing and collaborative editing functions of the report. The evaluation results need to be presented to relevant personnel in an intuitive and easy-to-understand way for analysis and decision-making. At the same time, highway construction and reconstruction involve multiple departments and require an efficient communication and collaboration mechanism. The traditional report generation and information transmission methods cannot meet the requirements. It enhances the visualization effect, facilitates users to intuitively understand the highway construction process and disease evolution, and is convenient for analysis and decision-making. It improves work efficiency, promotes multi-department cooperation, provides more detailed basis for subsequent decision-making in highway construction and reconstruction, and makes the whole process smoother and more efficient.

[0131] Data expansion applications assist scientific research and planning

[0132] Forms of expression: The combined data of material composition and environmental data accumulated over a long period of time for newly built roads can be used to analyze the variation law of material properties and provide a basis for material research and development. The correlation analysis of old road disease data and traffic flow data can provide a reference for preventive maintenance. In addition, infrared thermal imaging temperature data, micro-displacement data monitored by LiDAR equipment, and SAR data of satellite remote sensing can be used to study the impact of highway temperature changes on pavement performance, formulate highway maintenance cycles, and study the long-term change trend of the geological environment around the highway, respectively, providing a macro reference for highway planning and design. The expanded application of these data provides rich materials and research directions for scientific research in highway-related fields, promoting the development of highway science and technology. It helps to consider various factors in advance during the highway planning and design stage, optimize the design scheme, improve the durability and adaptability of the highway, and also provide a scientific basis for long-term maintenance and management.

[0133] The technical scalability drives the development of the industry

[0134] Technologies such as data fusion algorithms, models based on adversarial learning mechanisms, and data preprocessing algorithms in the system can be extended to other fields or similar engineering project evaluation systems, such as urban comprehensive monitoring systems and the identification of different types of building structures. The user feedback mechanism promotes the optimization of the evaluation index library, contributing to the formation of a consensus on evaluation standards within the industry. It promotes technical exchanges and integration between different fields, improving the technical level of related fields. It drives the entire highway construction and reconstruction industry towards standardization and scientific development, enhances the overall competitiveness of the industry, provides reference for other similar industries, and is conducive to the sustainable development of the industry.

[0135] Education and training resources enrich talent cultivation

[0136] Construction process animations can be used for the training of construction personnel to help new employees quickly understand the construction process and key technical links. Disease development history animations can be used as educational materials for studying the formation mechanism of diseases for researchers to analyze and summarize the development laws of diseases. The application experience of AR technology in the highway field can be extended to on-site monitoring in other engineering fields. It provides rich educational resources for the cultivation of talents in the highway industry, helps improve the skill level and professional quality of construction personnel, and trains more researchers who understand the mechanism of highway diseases. At the same time, it provides reference for the development of on-site monitoring technology in other engineering fields, promotes the overall technology dissemination and talent development in the engineering field, and is conducive to improving the quality of talents in the entire engineering field.

Claims

1. A highway traffic engineering test and detection data acquisition and evaluation system, characterized in that, It includes a data acquisition module, a data processing and analysis module, an evaluation index matching module, a data evaluation module, and a result output module: The data acquisition module includes an in-vehicle sensor unit. The in-vehicle sensor unit includes a material composition detection sensor for detecting the composition ratio of newly paved road materials, and a pavement structure layer detection radar for detecting the thickness change and internal disease conditions of the old road structure layer; The data processing algorithm of the in-vehicle data acquisition terminal analyzes the deviation between the newly built road material composition data and the preset design value in real time and gives adjustment suggestions, and uses the old road structure layer detection radar data combined with machine learning algorithms to identify the disease types and degrees; The data processing and analysis module includes a data cleaning and fusion unit and a feature extraction and road type recognition unit. For the newly built road data, a method based on the construction process model is used to identify abnormal data caused by construction process fluctuations. For the old road reconstruction data, combined with historical disease data and current environmental data, abnormal data caused by environmental changes is identified; Feature extraction and road type recognition unit: For newly built roads, construction process features are extracted, including but not limited to the vibration frequency features of construction machinery and the construction material transportation path features. For old road reconstruction, features related to the disease history are extracted, including but not limited to the first appearance time of the disease and the disease development speed; The evaluation index matching module includes an evaluation index library management unit and an automatic matching unit. The evaluation index library management unit, for newly built roads, the evaluation index library includes the impact indexes of construction noise and dust on surrounding residents and the ecological environment. For old road reconstruction, it includes traffic diversion ability evaluation indexes, including but not limited to temporary traffic control time and traffic congestion index; The automatic matching unit, when matching evaluation indexes and weights, considers construction progress and expected service life factors for newly built roads, and combines the historical evaluation data of the old road and the priority of the current reconstruction goal for old road reconstruction; The data evaluation module includes a single index evaluation unit and a comprehensive evaluation unit; The single index evaluation unit establishes an evaluation method that combines a special laboratory simulation model and on-site collected data for complex indexes of newly built roads, and introduces non-destructive testing technology for disease repair indexes in old road reconstruction; The comprehensive evaluation unit considers social impact factors in the process of newly built roads and old road reconstruction in the analytic hierarchy process, and optimizes the fuzzy set division method in fuzzy comprehensive evaluation; Analyze the sensitivity of each evaluation index to the comprehensive result; The result output module includes a visualization display unit and a report generation unit.

2. A highway traffic engineering test and detection data acquisition and evaluation system according to claim 1, characterized in that: The data acquisition module also includes an unmanned aerial vehicle (UAV) remote sensing unit: The UAV is equipped with an optical camera with infrared thermal imaging function and a LiDAR device with micro displacement monitoring ability, and is equipped with a professional UAV takeoff and landing platform and a control device that can realize automatic flight control; And an algorithm for planning flight routes according to the key areas of newly built road construction, the distribution of newly paved roads and old road diseases, and the key areas of reconstruction is introduced, which is used to conduct regular inspections on the road and transmit the collected image data and point cloud data back to the receiving device of the collection vehicle in real time.

3. A highway traffic engineering test and detection data acquisition and evaluation system according to claim 1, characterized in that: The data acquisition module further includes a satellite remote sensing data receiving unit: including a receiving antenna and a decoding device for receiving synthetic aperture radar data, as well as data preprocessing software; The data preprocessing software automatically identifies and removes the atmospheric interference and topographic shadow effects in the satellite remote sensing data by using artificial intelligence algorithms.

4. A highway traffic engineering test and detection data acquisition and evaluation system according to claim 1, characterized in that: The vehicle-mounted sensors are connected to the vehicle-mounted data acquisition terminal through the in-vehicle wiring system. The vehicle-mounted data acquisition terminal includes a data preprocessing subunit and a data compression subunit.

5. A highway traffic engineering test and detection data acquisition and evaluation system according to claim 1, characterized in that: In the data cleaning and fusion unit, a multi-source data fusion algorithm based on deep learning is adopted to control the fusion accuracy of data from different sources in space and time, assign high-level weights to the real-time data in the construction of new roads, and weight the long-term historical data in the renovation of old roads.

6. A highway traffic engineering test and detection data acquisition and evaluation system according to claim 1, characterized in that: For the feature extraction and road type recognition unit, an adversarial learning mechanism for improving the recognition ability of complex features in the construction of new roads and the renovation of old roads is introduced during the training of the machine learning model.

7. The highway traffic engineering test and detection data acquisition and evaluation system according to claim 1, characterized in that: In the automatic matching unit, an index matching assistance mechanism based on risk assessment is introduced to dynamically adjust the indexes and weights according to the road environment and abnormal situations during the data acquisition process.

8. The highway traffic engineering test and detection data acquisition and evaluation system according to claim 1, characterized in that: In the single index evaluation unit, an index evaluation subunit for analyzing the changing trend of the data indexes of new road construction and old road renovation over time is introduced.

Images