Data processing method of mountainous area highway road test signal detection system

Abstract

The invention discloses a data processing method of a mountainous area highway road test signal detection system, and relates to the technical field of data acquisition and processing, and the method comprises the following steps: S1, data standardization; s2, a data association strategy: performing association analysis on the standardized data by adopting a predetermined association rule and method, and establishing a logical relationship between the data to reveal an internal relationship between a road condition and a detection signal; s3, selecting and applying a data fusion algorithm: selecting a proper data fusion algorithm according to data characteristics and analysis requirements, and performing fusion processing on the associated multi-source data; s4, updating and feeding back a mechanism in real time; the method further comprises the steps that a unified data management platform is established and used for integrating and cooperatively processing all data involved in the steps S1 to S4; the mountainous area expressway road condition monitoring method is good in mountainous area expressway road condition monitoring effect and high in efficiency, and comprehensive assessment of the mountainous area expressway road condition is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of data acquisition and processing, and in particular to a data processing method for a mountain highway road test signal detection system. Background Art

[0002] With the development of mountain highway construction, accurate monitoring of road conditions is essential to ensure traffic safety and normal road operation. Mountain highways are located in a complex environment, and road conditions are affected by a variety of factors, including topography, geological conditions, climate, and frequent vehicle loads. In this case, an efficient and accurate data processing method for road test signal detection system is needed.

[0003] There are many deficiencies in the traditional road detection data processing method. On the one hand, the data sources are wide and the formats are diverse. The lack of unified standardized processing makes it difficult to effectively compare and comprehensively analyze data from different sources. For example, the data units and formats collected by different sensors are different, which brings great difficulties to subsequent processing. On the other hand, in terms of data association, it mostly relies on simple rules or manual experience, and it is impossible to accurately explore the deep internal connections between data, and it is difficult to fully reflect the complex relationship between road conditions and detection signals. At the same time, the fusion processing of multi-source data is not scientific enough, and the data characteristics and analysis requirements are not fully considered, which makes the final generated road condition assessment information inaccurate. Moreover, the traditional method also has defects in the timeliness of data update and feedback mechanism. The detection results cannot be transmitted to relevant departments in real time, resulting in the inability to respond to road problems in a timely manner. In addition, there is a lack of an effective data management platform, which makes it impossible to integrate and coordinate the entire data processing process, and there is no suitable visualization method, which is not conducive to users' understanding and analysis of data. These problems have seriously affected the effect and efficiency of road condition monitoring of mountain highways, and a new data processing method is urgently needed to solve these problems.

[0004] In view of this, this application is hereby filed. Summary of the invention

[0005] The object of the present invention is to provide a data processing method for a mountain highway road test signal detection system to solve the problems of poor effect and low efficiency of mountain highway road condition monitoring proposed in the above background technology.

[0006] To solve the above technical problems, a data processing method for a mountain highway road test signal detection system provided by the present invention includes the following steps: S1: Data standardization: Standardize the original data obtained from the mountain highway road test signal detection system to ensure that data from different sources and in different formats can be uniformly expressed and compared; S2: Data association strategy: Adopt predetermined association rules and methods to perform association analysis on the standardized data, establish the logical relationship between the data, and reveal the internal connection between the road conditions and the detection signals; S3: Data fusion algorithm selection and application: Select an appropriate data fusion algorithm according to the data characteristics and analysis requirements, and perform fusion processing on the associated multi-source data to generate more accurate and comprehensive road condition assessment information; S4: Real-time update and feedback mechanism: Establish a real-time data update system to ensure the timeliness and accuracy of the detection data, and through the feedback mechanism, timely transmit the processing results to the relevant road maintenance and management departments for timely response measures; It also includes: establishing a unified data management platform for integrating and collaboratively processing all the data involved in steps S1 to S4 to achieve a comprehensive assessment of the mountain highway road conditions.

[0007] Further, in S1, it also includes identifying and processing abnormal data to improve data quality.

[0008] Further, in S2, an association method based on timestamp, geographical location or signal characteristics is adopted to achieve precise matching and association of the data.

[0009] Further, in S2, it also includes a data association model based on machine learning or deep learning algorithms, which can automatically identify and learn the potential association patterns between the data, further improving the accuracy and efficiency of data association analysis.

[0010] Further, in S3, according to the characteristics and analysis requirements of the data, one or more data fusion algorithms such as weighted average method, Kalman filtering method, Bayesian network method or neural network method are selected for data processing.

[0011] Further, in S4, it also includes establishing a data quality monitoring system to monitor the data update and feedback process in real time to ensure the accuracy and timeliness of the data.

[0012] Further, the data management platform in S1 also includes a data visualization module for intuitively displaying the processed data in the form of charts, reports, etc., facilitating user understanding and analysis.

[0013] Further, in S1, there is also an adaptive adjustment module, which can dynamically adjust the parameters and settings in data standardization, association strategy, data fusion algorithm, and real-time update and feedback mechanism according to the actual operation conditions and data changes of the road test signal detection system, so as to optimize the data processing flow and improve the flexibility and adaptability of data processing.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0015] 1. By adopting predetermined association rules and methods, such as association methods based on timestamps, geographical locations or signal characteristics, and combining with a data association model of machine learning or deep learning algorithms, potential association patterns between data can be automatically identified and learned. This not only improves the accuracy of data association analysis, but also greatly enhances the processing efficiency, accurately reveals the internal connections between data, and provides strong support for comprehensively evaluating road conditions;

[0016] 2. For different types of data, appropriate algorithms can be selected for fusion according to their characteristics. The weighted average method can comprehensively consider the importance of each data for evaluation, the Kalman filtering method can effectively eliminate noise in the data, the Bayesian network method can establish the probability relationship between relevant data, and the neural network method can process complex non-linear relationships, thereby generating more accurate and comprehensive road condition evaluation information;

[0017] 3. By establishing a unified data management platform, integrating and collaboratively processing data in each link, a comprehensive evaluation of the road conditions of mountain expressways can be achieved. This platform can effectively coordinate operations such as data standardization, association, fusion, update, and feedback, and improve the integrity and coherence of data processing.

[0018] 4. The data visualization module visually displays the processed data in the form of charts, reports, etc., such as drawing line charts of displacement changing with time, bar charts of stress distribution, etc. It is convenient for users to understand and analyze the data, enabling managers and technicians to quickly obtain key information of the data, providing convenience for decision-making, and improving work efficiency.

[0019] 5. The adaptive adjustment module can dynamically adjust the parameters and settings in the data processing flow according to the actual operation conditions and data changes of the road test signal detection system, including aspects such as data standardization, association strategy, data fusion algorithm, and real-time update and feedback mechanism. This flexibility enables the system to adapt to the complex and changeable environment of mountain expressways and different operation stages, always maintaining the high efficiency and accuracy of data processing, and reducing the maintenance cost and management difficulty. Brief Description of the Drawings

[0020] Figure 1 It is a principle block diagram of a data processing method for a road test signal detection system of a mountain expressway. Specific implementation manners

[0021] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to 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.

[0022] Please refer to Figure 1 , the present invention provides a technical solution: a data processing method for a road test signal detection system of a mountain expressway, including the following steps:

[0023] S1: Data standardization: Standardize the original data obtained from the road test signal detection system of the mountain expressway to ensure that data from different sources and in different formats can be uniformly expressed and compared. It also includes identifying and processing abnormal data to improve data quality;

[0024] S2: Data association strategy: Adopt predetermined association rules and methods to perform association analysis on the standardized data, establish logical relationships between the data, to reveal the internal connection between the road conditions and the detection signals. Adopt association methods based on timestamps, geographical locations or signal characteristics to achieve precise matching and association of the data. It also includes a data association model based on machine learning or deep learning algorithms, which can automatically identify and learn the potential association patterns between the data, further improving the accuracy and efficiency of data association analysis;

[0025] S3: Selection and application of data fusion algorithms: According to the data characteristics and analysis requirements, select appropriate data fusion algorithms to fuse the associated multi-source data to generate more accurate and comprehensive road condition evaluation information. According to the characteristics and analysis requirements of the data, select one or more data fusion algorithms such as the weighted average method, the Kalman filtering method, the Bayesian network method or the neural network method for data processing;

[0026] S4: Real-time update and feedback mechanism: Establish a real-time data update system to ensure the timeliness and accuracy of the detection data, and through the feedback mechanism, timely transmit the processing results to the relevant road maintenance and management departments for timely response measures. It also includes establishing a data quality monitoring system to monitor the data update and feedback process in real time to ensure the accuracy and timeliness of the data;

[0027] It also includes: establishing a unified data management platform for integrating and collaboratively processing all the data involved in steps S1 to S4 to achieve a comprehensive assessment of the road conditions of mountain expressways. It also includes a data visualization module for intuitively displaying the processed data in the form of charts, reports, etc., facilitating user understanding and analysis.

[0028] It also includes an adaptive adjustment module that can dynamically adjust the parameters and settings in data standardization, correlation strategies, data fusion algorithms, and real-time update and feedback mechanisms according to the actual operation conditions and data changes of the road test signal detection system to optimize the data processing process and improve the flexibility and adaptability of data processing.

[0029] Example 1: Applied to the data processing of the road tunnel conditions of mountain expressways:

[0030] S1: Data collection and standardization:

[0031] Install various sensors at different key positions in the tunnel, such as displacement sensors (for deformation monitoring), stress-strain sensors (for stress / strain monitoring), crack detection instruments, leakage detection equipment, environmental monitoring equipment, support structure detection equipment (such as bolt pull-out test equipment), and concrete quality detection equipment (such as rebound hammers), etc. These sensors collect data at a certain frequency, including displacement data of each measuring point in the tunnel (such as the settlement data of a certain measuring point per week), stress data (such as concrete strain, steel bar stress data), crack information (width, length), leakage-related data (location, water volume), environmental data (oxygen content, temperature), support structure parameters (bolt tensile strength), and concrete quality data (strength, carbonation depth), etc.

[0032] Perform standardization processing on the collected original data. For example, uniformly convert the displacement data to millimeters and the stress data to megapascals, and at the same time, dynamically adjust the standardization parameters through the adaptive adjustment module according to the actual operation conditions and data changes of the tunnel. If it is found that the fluctuation range of the displacement data increases in a certain stage, the standardization scale can be adjusted accordingly. In this process, identify and process abnormal data. For example, if the displacement sensor has data with extremely large values in a short period of time and does not conform to the normal deformation law of the tunnel, it may be a sensor failure, and such data will be marked or corrected. The data visualization module of the data management platform displays some of the standardized data in the form of charts, such as drawing a line chart of displacement change over time, a bar chart of stress distribution, etc., for intuitive viewing.

[0033] S2: Data correlation:

[0034] Adopt an association method based on timestamps, geographical locations, or signal characteristics. For example, associate the displacement data and stress data at a certain location collected at the same time, and analyze the force and deformation relationship of the tunnel structure at this location at this moment. At the same time, use a data association model based on machine learning or deep learning algorithms. This model automatically identifies and learns the potential association patterns between data through learning a large amount of historical tunnel monitoring data. For example, the model can learn the association between the water leakage location and the development of lining cracks, further improving the accuracy and efficiency of data association analysis.

[0035] S3: Data fusion:

[0036] Select a suitable data fusion algorithm according to the characteristics of the data and the analysis requirements. For multi-source data such as displacement and stress, if the data is relatively stable and the weights are clear, the weighted average method can be selected for fusion; if the data has certain dynamic changes and uncertainties, the Kalman filtering method can be used. For example, when evaluating the overall stability of the tunnel structure, the displacement data and stress data of different measuring points are fused by the weighted average method to generate more accurate and comprehensive tunnel structure condition assessment information. For the strength data and carbonation depth data in concrete quality inspection, different weights can be assigned according to their importance to structural safety and then fused.

[0037] S4: Real-time update and feedback:

[0038] Establish a real-time data update system. The sensors continuously collect and upload data to ensure the timeliness and accuracy of the detection data. At the same time, establish a data quality monitoring system to monitor the data update and feedback process in real time. For example, when it is found that the data uploaded by a certain sensor remains unchanged for a long time or the data format is incorrect, an alarm is issued in time and measures are taken.

[0039] Transmit the processing results to the tunnel maintenance and management department in a timely manner through a feedback mechanism. If it is found that the width of the tunnel lining crack exceeds the safety threshold, the water leakage seriously affects the structural safety, the stability of the support structure is insufficient, or the environmental indicators are abnormal, etc., the relevant departments shall be notified in time.

[0040] Specifically:

[0041] I: Deformation monitoring data:

[0042] In the long-term monitoring of a tunnel, deformation monitoring data is the key basis for evaluating the stability of the tunnel structure. For the clearance convergence data of a specific measuring point in a certain tunnel, its changes show a certain pattern: it was 0.2 mm in the first week. This tiny change may not be easily noticed in the initial stage, but as time goes by, it reached 0.3 mm in the third week and further increased to 0.4 mm in the sixth week. This gradually increasing trend indicates that the tunnel structure in the area where this measuring point is located may be under continuous pressure or affected by external factors.

[0043] The crown settlement is also an indicator that cannot be ignored. Within one month, the crown settlement increased from 1 mm to 1.5 mm. This change range reflects that there may be potential safety hazards in the crown structure. The surrounding rock settlement and soil horizontal / vertical displacement data are also collected and recorded at certain time intervals, such as once a week or once every two weeks. These data depict in detail the movement state of the surrounding rock and soil, providing detailed information for analyzing the influence of the surrounding geological environment on the tunnel structure. For example, if the settlement speed of the surrounding rock in a certain area suddenly accelerates, it may indicate that there are potential geological hazard risks in this area, such as formation loosening or abnormal groundwater activities, etc.

[0044] II: Stress / strain monitoring data:

[0045] Stress / strain monitoring data is crucial for deeply understanding the performance of the tunnel structure under the stress state. Taking the concrete strain data as an example, the strain values of the concrete at a certain cross-section at different stages clearly reflect the stress changes of the structure. The strain value was 30 με one month after construction, and at this time the structure was in the initial stress adjustment stage; while it was 50 με at three months. The increase in the strain value means that the load borne by the structure is continuously accumulating or the stress distribution has changed.

[0046] As the main load-bearing component, the stress data of the steel bars can more intuitively reflect the stress condition of the tunnel structure. The stress values of the steel bars at key parts are significantly different under different loads. For example, the stress of a main load-bearing steel bar is 100 MPa under normal operating loads, which indicates that the structure is in a relatively stable stress state under daily operating conditions. However, the stress reaches 200 MPa under special working conditions (such as earthquake simulation). This large-scale stress change highlights the huge impact of special situations on the structure and also provides key data for evaluating the bearing capacity of the tunnel under extreme conditions.

[0047] In addition, pore water pressure and supporting soil pressure data are also accurately collected at corresponding positions and time points. The change of pore water pressure may affect the stability of the soil mass and the stress condition of the structure. For example, when the pore water pressure increases, it may lead to a decrease in the shear strength of the soil mass, thus exerting a greater lateral pressure on the supporting structure. The supporting soil pressure data directly reflect the external loads borne by the supporting structure and are of great significance for optimizing the supporting design and evaluating the safety of the supporting structure.

[0048] III: Crack detection data:

[0049] Cracks are important warning signals for the safety of tunnel structures. Their detailed detection and data recording are crucial for tunnel maintenance. The initial width of a certain crack is 0.1 mm and the length is 5 cm. This tiny crack may have a relatively small impact on structural safety in the initial stage, but after two months, the width becomes 0.2 mm and the length increases to 8 cm. This obvious development trend indicates that the crack is in an active development stage and may have a serious impact on the integrity and stability of the tunnel structure.

[0050] Throughout the tunnel, data on cracks at different positions are recorded in detail, including their development speed and change trend. For example, the development speed of cracks at the waist of the tunnel arch may be faster than that at the crown, which may be related to factors such as the stress characteristics, geological conditions, or construction quality at the waist of the arch. Through long-term monitoring and data analysis, a crack development model can be established to predict the future development direction of the crack and the possible degree of damage to the structure, thus providing a basis for timely taking repair measures.

[0051] IV: Lining quality detection data:

[0052] As the inner protective structure of the tunnel, the quality of the lining is directly related to the durability and safety of the tunnel. The lining thickness data obtained through advanced ultrasonic detection technology provide an intuitive basis for evaluating the lining quality. For example, the average thickness of a certain section of the lining is 35 cm. Whether this thickness meets the design requirements and the thickness variation in different areas need further analysis.

[0053] At the same time, detecting the size of the cavity behind the lining (such as the cavity area is 0.5 m 2 ) and the position information (200 m from the tunnel entrance, 2 m high) is equally crucial. The cavity behind the lining may cause uneven stress on the lining structure and local stress concentration, thus accelerating the damage of the lining. The existence of these cavities may be due to incomplete grouting during construction or geological condition changes. Precise detection of them helps to formulate targeted repair plans, such as filling the cavities through grouting and other methods to restore the integrity of the lining structure.

[0054] V: Leakage detection data:

[0055] The problem of seepage water is a common and harmful problem in tunnel operation, and its relevant detection data is crucial for evaluating the waterproof performance and structural safety of the tunnel. The accurate recording of the seepage water location coordinates (such as 300m from the tunnel entrance and 2.5m in height) helps to quickly locate the leakage point and take repair measures in a timely manner. The seepage water volume (15ml leakage per minute) reflects the severity of the leakage, and different water volumes have different impacts on the tunnel structure and equipment.

[0056] The turbidity level (turbidity value is 20NTU) can provide information about the source of the seepage water and the possible substances carried. If the turbidity of the seepage water is high, it may contain impurities such as sediment, which will not only affect the environmental hygiene in the tunnel but also may cause blockage of the drainage system. The freezing condition (icing occurs when the temperature is below -2°C) is an issue that cannot be ignored. Icing may cause damage to the lining structure due to frost heave, affecting the normal use of the tunnel.

[0057] The blockage situation of the original waterproof and drainage system (the blockage rate is determined to be 20% through pipeline flow detection) is directly related to the treatment effect of the seepage water problem. If the waterproof and drainage system is severely blocked, even if the leakage point is repaired, the seepage water problem may not be effectively solved. Therefore, it is necessary to regularly inspect and maintain the waterproof and drainage system.

[0058] VI: Detection data of the support structure:

[0059] The support structure is a key part to ensure the safety during tunnel excavation and operation, and its detection data is of great significance for evaluating the support effect and structural stability. In the bolt pull-out test, the tensile strength of the bolt (such as the average tensile strength is 180MPa) is an important indicator to measure the quality and support capacity of the bolt. If the tensile strength of the bolt is insufficient, it may not be able to effectively fix the surrounding rock, resulting in loosening of the surrounding rock and tunnel deformation.

[0060] The deformation data of the steel arch (the displacement value under stress, such as the displacement of a certain node is 3mm) reflects the working state of the steel arch when bearing external loads. Excessive displacement may indicate problems in the design or installation of the steel arch, or the load borne exceeds its design bearing capacity, and it is necessary to adjust or reinforce it in a timely manner.

[0061] The strength data of the shotcrete (the value obtained through the compressive test, such as the average strength is C30) is equally crucial. As part of the support structure, the strength of the shotcrete directly affects the support effect. Insufficient strength may cause cracks or spalling of the concrete during stress, reducing the overall stability of the support structure.

[0062] VII: Detection data of concrete quality:

[0063] The quality of concrete is one of the core elements for the safety of tunnel structures. The data obtained through various detection means comprehensively reflects the performance of concrete. The concrete strength data obtained by a rebound hammer (such as the strength grades corresponding to the rebound values at different positions are between C25 - C35) shows the strength differences of concrete at different locations. Such differences may be caused by different construction techniques, raw material qualities, or curing conditions, and further analysis is required to evaluate the overall strength of the concrete structure.

[0064] The carbonation depth (the value obtained by detecting with phenolphthalein reagent, such as the average carbonation depth is 6 mm) is an important indicator to measure the durability of concrete. Carbonation reduces the alkalinity of concrete, destroys the passivation film on the surface of steel bars, and thus accelerates the corrosion of steel bars. When the carbonation depth exceeds a certain limit, protective measures such as surface coatings need to be taken to protect the steel bars and the concrete structure.

[0065] In addition, durability - related indicators (such as data on chloride ion penetration resistance) are crucial for evaluating the long - term performance of concrete in complex environments. In environments with a risk of chloride ion erosion, such as road tunnels near the sea or using de - icing salts, the chloride ion penetration resistance of concrete directly affects its service life. Low chloride ion penetration resistance may lead to the corrosion of steel bars and the damage of the concrete structure, so corresponding protective measures need to be taken according to the test results.

[0066] VIII: Environmental monitoring data:

[0067] The environmental monitoring data in the tunnel is of great significance for ensuring the safety and comfort of personnel in the tunnel and the normal operation of equipment. The oxygen content in the tunnel (such as 19%) is close to the lower limit of normal air oxygen content and needs to be closely monitored to ensure that personnel in the tunnel do not have safety problems such as difficulty in breathing due to lack of oxygen.

[0068] Although the concentration of harmful gases (carbon monoxide concentration is 8 ppm) is currently at a low level, in special situations such as vehicle exhaust emissions or fires, the concentration of harmful gases may rise rapidly. Therefore, it is necessary to monitor in real - time and equip with an effective ventilation system to discharge in time when the concentration of harmful gases exceeds the standard.

[0069] An environment with a relatively high humidity (75%) may promote the growth of mold and metal corrosion, causing damage to the equipment and structure in the tunnel. The temperature (18°C) affects the comfort of personnel and the operation of some temperature - sensitive equipment. Through the analysis of environmental monitoring data, the operating parameters of ventilation, dehumidification and other equipment can be reasonably adjusted to create a safe and comfortable tunnel operation environment.

[0070] It should be noted here that:

[0071] I: In S2, the embodiment of the data association strategy:

[0072] 1. Association between deformation monitoring data and other data:

[0073] Association with stress / strain data: The change trends of deformation data such as convergence and crown settlement are related to concrete strain and steel bar stress data. For example, when the convergence and crown settlement increase, such as the crown settlement increasing from 1 mm to 1.5 mm, the stress / strain of the corresponding structural part will also change, like the concrete strain value increasing from 30 με to 50 με. This indicates that the deformation may be caused by the change of structural stress, and the two confirm each other, jointly reflecting the stress state and stability change of the tunnel structure.

[0074] Association with crack detection data: Deformation data can be associated with the development of cracks. With the increase of deformation, such as the continuous increase of convergence at a certain measuring point, it may lead to the appearance of cracks or the expansion of existing cracks. For example, a crack with an initial width of 0.1 mm and a length of 5 cm becomes 0.2 mm in width and 8 cm in length under the continuous action of deformation, indicating that deformation is an important influencing factor for crack development.

[0075] Association with lining quality detection data: Larger deformation may affect the lining quality. When deformations such as convergence and crown settlement occur in the tunnel, it may cause the cavity behind the lining to expand or lead to uneven stress on the lining, affecting the change of lining thickness. For example, in a certain section with an average lining thickness of 35 cm, the thickness may change in local areas due to deformation, affecting the integrity and protection effect of the lining.

[0076] 2. Association between stress / strain monitoring data and other data:

[0077] Association with deformation monitoring data: The change of stress / strain data can explain the phenomena in deformation monitoring data. For example, when the steel bar stress increases from 100 MPa to 200 MPa under special working conditions, this stress change may lead to more severe deformation of the tunnel structure, which is correlated with the changes of deformation data such as convergence and crown settlement, helping to analyze the mechanical reasons for structural deformation.

[0078] Association with support structure detection data: Stress / strain data is related to the mechanical properties of the support structure. When the concrete strain or steel bar stress changes, it will affect the stress of support structures such as bolts, steel arches, and shotcrete. For example, an increase in stress may cause bolts to bear greater tensile force. If the bolt tensile strength is insufficient (such as an average tensile strength of 180 MPa), it may not be able to resist this change, resulting in surrounding rock loosening, and then affecting the stability of the entire support structure and the tunnel.

[0079] Association with concrete quality inspection data: The stress / strain changes are closely related to the concrete quality. Quality problems such as insufficient concrete strength or increased carbonation depth will affect its stress / strain performance. For example, if the concrete strength fluctuates between C25 - C35 and the average carbonation depth is 6mm, it will change the mechanical properties of the concrete, and further affect the stress / strain state of the structure. The association between the two can analyze the influence of concrete quality on the structural stress.

[0080] 3. Association between crack detection data and other data:

[0081] Association with deformation monitoring data and stress / strain data: The development of cracks interacts with deformation and stress / strain. Changes in deformation and stress will promote the generation and development of cracks, and the appearance and expansion of cracks will change the stress state of the structure, further affecting the deformation. For example, the crack development speed at the arch waist position is faster than that at the arch top, which may be related to the deformation and stress characteristics of this part. Through this association, the safety of the structure can be evaluated more comprehensively.

[0082] Association with lining quality inspection data: The existence and development of cracks will have a negative impact on the lining quality. Problems such as voids behind the lining may cause cracks in the lining structure. At the same time, the expansion of cracks will further damage the integrity of the lining and reduce its protection function. For example, when the area of the void behind the lining is 0.5m 2 it may cause cracks in the nearby lining and accelerate their development.

[0083] 4. Association between lining quality inspection data and other data:

[0084] Association with leakage detection data: Lining quality problems such as insufficient thickness or voids behind may lead to leakage. When there are voids in the lining or the thickness does not meet the requirements, leakage is likely to occur. For example, at a location 200m from the tunnel entrance and 2m high where there is a void in the lining, it may cause leakage problems in this area. The association between the leakage location and the lining quality problems helps to determine the cause of the leakage.

[0085] Association with support structure inspection data: The lining quality and the support structure interact with each other. The performance of the support structure affects the stress condition of the lining. If there are problems with the support structure such as bolts, steel arch frames or shotcrete, it may cause uneven stress on the lining and affect the lining quality. For example, insufficient tensile strength of bolts or excessive displacement of steel arch frames will change the load distribution borne by the lining, and further affect the thickness and integrity of the lining.

[0086] 5. Association between leakage detection data and other data:

[0087] Association with lining quality inspection data: The problem of leakage is closely related to the lining quality. Poor lining quality is one of the common causes of leakage. For example, voids behind the lining or insufficient thickness can lead to leakage. At the same time, leakage will further erode the lining, reducing the lining quality and forming a vicious cycle. The association between the two can analyze the root cause of leakage and its impact on the structure.

[0088] Association with environmental monitoring data: The turbidity, freezing condition, etc. of leakage are related to environmental conditions. When the humidity in the tunnel is 75% and the temperature is below -2°C under special circumstances, the leakage may freeze, affecting the lining structure. High turbidity of leakage may be related to environmental dust and impurities in the tunnel and is also affected by environmental factors such as ventilation.

[0089] 6. Association between support structure inspection data and other data:

[0090] Association with deformation monitoring data: The performance of the support structure directly affects the tunnel deformation. Problems such as insufficient tensile strength of bolts, excessive displacement of steel arches, or insufficient strength of shotcrete will lead to increased tunnel deformation. By associating with deformation data such as convergence of the clear span and settlement of the crown, the effect of the support structure on controlling deformation can be evaluated, and the deficiencies of the support structure can be detected in a timely manner.

[0091] Association with concrete quality inspection data: Data such as the strength of shotcrete in the support structure is related to the concrete quality. The average strength of shotcrete is C30. If its quality does not meet the requirements, such as large strength fluctuations or poor durability, it will affect the stability of the support structure. The association with concrete quality inspection data can analyze the reliability of the support structure.

[0092] 7. Association between concrete quality inspection data and other data:

[0093] Association with stress / strain monitoring data: The concrete quality determines its stress / strain performance. Quality conditions such as the concrete strength being between C25 - C35 and the carbonation depth being 6mm will affect the strain and stress changes of the structure under load. The association between the two can evaluate the impact of concrete quality on the mechanical properties of the structure.

[0094] Association with environmental monitoring data: The concrete quality is affected by environmental factors and also affects the environment. For example, in an environment with a humidity of 75%, a temperature of 18°C, and a risk of chloride ion erosion, the durability indicators such as the chloride ion penetration resistance of the concrete interact with the environmental conditions. Poor-quality concrete is more likely to be damaged in such an environment, affecting the safety of the tunnel structure and the environmental conditions.

[0095] 8. Association between environmental monitoring data and other data:

[0096] Associated with the data of seepage water detection: Conditions such as environmental humidity and temperature affect the state of seepage water. High humidity may make the seepage water problem more serious, low temperature may cause the seepage water to freeze, and the presence of seepage water will in turn change conditions such as the local environmental humidity. The two affect each other and jointly affect the operation safety of the tunnel.

[0097] Associated with various other types of data: Factors such as the oxygen content and harmful gas concentration in the environment will affect the maintenance of the tunnel structure and equipment by personnel. When the oxygen content is 19% or the carbon monoxide concentration is 8 ppm, personnel need to consider environmental safety when inspecting and maintaining the structure and equipment. At the same time, the condition of the structure and equipment will also have a certain impact on the environment. For example, seepage water may affect the air humidity and harmful gas concentration.

[0098] II. In S3, the embodiment of the selection and application of the data fusion algorithm:

[0099] 1. Embodiment of the application of the weighted average method:

[0100] Deformation monitoring data:

[0101] For different types of deformation monitoring data such as clearance convergence data, crown settlement data, surrounding rock settlement and soil displacement data, different weights can be assigned according to their importance in affecting the stability of the tunnel structure, and then the weighted average method can be used. For example, if it is considered that the crown settlement data is more important for the structural safety assessment than the surrounding rock settlement data, a higher weight can be given to the crown settlement data. Suppose the weight of the crown settlement is 0.6 and the weight of the surrounding rock settlement is 0.4. When the crown settlement amount is 1.2 mm and the surrounding rock settlement amount is 0.5 mm at a certain moment, the weighted average deformation amount is 1.2×0.6 + 0.5×0.4 = 0.88 mm. In this way, the comprehensive influence of various deformation data on the stability of the tunnel structure can be considered.

[0102] Stress / strain monitoring data:

[0103] When fusing concrete strain data and steel bar stress data, since the steel bar stress is more intuitively representative of the structural stress condition, a higher weight can be assigned. For example, the weight of the steel bar stress is 0.7 and the weight of the concrete strain is 0.3. When the construction is 3 months old, the steel bar stress is 150 MPa and the concrete strain is 40 με. After appropriate unit conversion and weighted average, a value that comprehensively reflects the structural stress can be obtained for more accurately evaluating the structural stress state.

[0104] 2. Embodiment of the application of the Kalman filter method:

[0105] Deformation monitoring data:

[0106] In long-term tunnel deformation monitoring, there may be noise in the measurement data. For example, during the acquisition of clearance convergence data, it may be interfered by factors such as measurement instrument errors and environmental vibrations. The Kalman filtering method can filter these noisy clearance convergence data to obtain a more accurate convergence trend. The same applies to the crown settlement data. Through Kalman filtering, the true change trend of the settlement can be better analyzed, and some short-term abnormal fluctuations can be removed, so as to more accurately judge whether there are potential safety hazards in the crown structure.

[0107] Stress / strain monitoring data:

[0108] For concrete strain and steel bar stress data, there may be data fluctuations during the acquisition process due to factors such as sensor accuracy and external electromagnetic interference. The Kalman filtering method can predict the data at the next moment based on the existing data sequence and correct it in combination with the new measurement values, so as to obtain a smoother and more accurate strain and stress change trend, which is more conducive to analyzing the force changes of the structure at different stages.

[0109] 3: Application manifestations of the Bayesian network method:

[0110] Crack detection data and lining quality detection data:

[0111] The development of cracks is closely related to the lining quality. The Bayesian network method can establish a probability relationship model between data such as crack width and length and lining quality data such as lining thickness and the size of voids behind the lining. For example, given that the width of a certain crack increases and the area of the void behind the lining is large, the probability of serious problems in the tunnel structure in this area can be inferred through the Bayesian network, providing a basis for maintenance decisions. At the same time, the correlation probability between the crack development speed at different positions and the quality problems in different areas of the lining can also be analyzed to better predict the crack development trend and lining damage conditions.

[0112] Leakage detection data and support structure detection data:

[0113] The leakage problem may affect the stability of the support structure. The Bayesian network can establish the relationship between data such as leakage location, water volume, and turbidity and support structure detection data such as the tensile strength of anchor bolts, the deformation of steel arch frames, and the strength of shotcrete. For example, when the leakage water volume is large and the turbidity is high, the probability of the tensile strength of the anchor bolts decreasing due to corrosion and the deformation of the steel arch frames increasing due to water erosion will increase. Through this probability model, the safety of the support structure under leakage conditions can be evaluated in advance.

[0114] 4: Application manifestations of the neural network method:

[0115] Comprehensive analysis of environmental monitoring data and other data:

[0116] The neural network method can handle the non-linear relationships between various complex data. For the environmental monitoring data such as oxygen content, harmful gas concentration, humidity, temperature, etc. inside the tunnel and other data such as deformation monitoring data, stress / strain monitoring data, etc., a neural network model can be established. For example, when the oxygen content decreases, the temperature rises and the crown settlement increases, the neural network can learn from a large amount of historical data the degree of risk that the tunnel structure may face in such a complex situation, and then comprehensively evaluate and predict the overall safety of the tunnel. At the same time, for the fusion analysis of multi-source data such as different types of crack data, lining quality data, leakage data, etc., the neural network can also discover the hidden patterns in them, providing a more accurate basis for tunnel maintenance and safety assessment.

[0117] Example 2: Application to the processing of road surface flatness detection data of mountain expressways:

[0118] Data standardization:

[0119] 1. Detailed data format processing:

[0120] For the road surface flatness detection data, in addition to unifying the units, it is also necessary to standardize the precision of the data. For example, if the flatness value output by the detection device is accurate to two decimal places, it is uniformly standardized to this precision form. At the same time, for the possible differences in coding formats of different devices, such as data in ASCII code or Unicode encoding, it is uniformly converted into a coding form suitable for system processing.

[0121] In-depth methods for abnormal data processing;

[0122] Data association strategy:

[0123] 1. Refinement of timestamp association

[0124] In addition to analyzing the changes in road surface flatness by season and daily time periods, it can be further subdivided into hours. For example, it is statistically found that during the period from 2 to 5 am every day, due to the reduction of vehicle load and relatively stable road surface temperature, the fluctuations of the road surface flatness data are relatively small at this time. During the morning and evening rush hours (7-9 am, 17-19 pm), vehicles frequently brake and accelerate, and the load impact is large, and the change in flatness may be more obvious. Collect the flatness data of a mountain expressway at different hours every day for a month. The standard deviation of flatness during the peak hours can reach 0.3 mm, while the standard deviation during the early morning low peak hours may be only 0.1 mm.

[0125] 2. Expansion of geographical location association:

[0126] Combined with the Geographic Information System (GIS), the pavement evenness data is accurate to the location of each one-meter road section. Special road sections such as bridges, tunnels, curves, and ramps are marked and analyzed separately. For example, due to the structural characteristics and vehicle driving vibrations of a certain bridge section, the change in evenness may be significantly different from that of ordinary road sections. Through data collection at multiple detection points at both ends and in the middle of the bridge, it is found that the average deviation of evenness in the middle is 0.2 mm larger than that at both ends.

[0127] 3. Training and application of the machine learning correlation model:

[0128] Collect data on various factors including pavement type (such as asphalt, cement), traffic flow (such as 5000 - 10000 vehicles per day), road service life (such as 5 - 10 years), and climate conditions (such as areas with an average annual rainfall of 800 - 1000 mm) as features, and train a machine learning model together with the pavement evenness data. For example, using the decision tree algorithm, after training with a large amount of data, it is found that for road sections with a service life of more than 8 years, asphalt pavement, and an annual rainfall of more than 900 mm, the rate of decline in evenness is about 10% faster than that of road sections with similar other conditions.

[0129] Data fusion algorithm selection and application:

[0130] 1. Basis for determining the weights of the weighted average method:

[0131] Determine the weights according to the historical accuracy of the detection equipment. Assume that the accuracy of equipment A in pavement evenness detection in the past year is 95%, and that of equipment B is 90%. Then, when calculating the weighted average, the weight of equipment A can be set to 0.6, and the weight of equipment B to 0.4. If the evenness value detected by equipment A is 3.5 mm and that by equipment B is 3.8 mm, the weighted average evenness value is 3.5 x 0.6 + 3.8 x 0.4 = 3.62 mm.

[0132] 2. Parameter setting and application example of the Kalman filtering method:

[0133] Set the parameters of the Kalman filter according to the dynamic characteristics of the change in pavement evenness. Assume that the change in pavement evenness in a short period of time conforms to a first-order Markov process, and the state transition matrix is set as a time-related function based on experience. For example, for a detection interval of every 10 minutes, the parameter in the state transition matrix is set to 0.9, indicating a strong correlation between the evenness at the current moment and the previous moment. The measurement noise covariance is determined according to the accuracy of the detection equipment. If the equipment accuracy is ±0.2 mm, the measurement noise covariance is set to 0.04. Through the Kalman filter, the change trend of pavement evenness can be predicted more accurately, such as predicting the change range of evenness in the next 10 minutes.

[0134] 3. Structure and parameter learning of the Bayesian network method:

[0135] Build a Bayesian network that includes nodes such as pavement materials, traffic flow, climate conditions, and pavement smoothness. Learn the structure and parameters of the network, i.e., the conditional probability table, through a large amount of historical data. For example, given that a certain road section is an asphalt pavement, the traffic flow is 8,000 vehicles per day, and the annual rainfall is 900 mm, the probability that the pavement smoothness is excellent (smoothness value less than 3 mm) is 70%, the probability that it is medium (smoothness value 3 - 5 mm) is 25%, and the probability that it is poor (smoothness value greater than 5 mm) is 5%.

[0136] 4. Architecture of the neural network method and preparation of training data:

[0137] Adopt a multi-layer perceptron neural network architecture. The input layer includes factors such as road service life, traffic flow, and climate temperature, and the output layer is the pavement smoothness value. Collect at least 1,000 sets of complete data under different road sections and conditions for training. For example, after training, the neural network outputs a predicted smoothness value of 3.2 mm for the input data of a new road section (service life of 6 years, traffic flow of 7,000 vehicles per day, and average annual temperature of 15°C).

[0138] Real-time update and feedback mechanism:

[0139] 1. Optimization of the real-time data update system:

[0140] Adopt 5G communication technology to achieve faster and more stable data transmission, ensuring that the pavement smoothness detection data is uploaded to the data management platform within 1 minute after collection. At the same time, adopt data encryption technology, such as the Advanced Encryption Standard (AES) algorithm, during data transmission to ensure data security.

[0141] 2. Diversification and timeliness of the feedback mechanism:

[0142] In addition to text messages, emails, and management software notifications, a mobile application can also be developed. When there are abnormal changes in the pavement smoothness data (such as the smoothness value changing by more than 0.5 mm within a day), relevant department personnel can immediately receive a push notification on the mobile device, and at the same time, a report containing detailed data and analysis suggestions is pushed. The report content can include the location of the abnormal road section, possible reasons (such as the roadbed being loosened due to recent heavy rain erosion), and preliminary maintenance suggestions (such as emergency inspections and local repairs).

[0143] 3. Specific measures for the data quality monitoring system:

[0144] During the data update and feedback process, in addition to checking the integrity and accuracy of data transmission, the timeliness of data can also be monitored. For example, set the maximum allowable delay time for each detection cycle. If data is not received beyond this time, trigger the alarm mechanism. At the same time, cross-verify the data processing results. Use different algorithms or models to process the same batch of data. If the results differ by more than a certain threshold (such as 10%), recheck the data and the processing flow.

[0145] Unified Data Management Platform and Data Visualization Module:

[0146] 1. Function Expansion of the Data Management Platform:

[0147] In the unified data management platform, establish a data warehouse to store and archive historical road surface flatness detection data for a long time. Adopt distributed storage technology, such as the Hadoop Distributed File System (HDFS), to ensure the high availability and scalability of data. At the same time, the platform has a data mining function, which can automatically discover potential patterns from a large amount of data. For example, find groups of road sections with similar flatness changes through clustering analysis, providing a reference for road maintenance planning.

[0148] 2. Rich Forms of Data Visualization:

[0149] In addition to common charts and reports, use 3D modeling technology to visualize road surface flatness data. For example, create a 3D model of the entire mountain highway, representing road sections with different flatness ranges in different colors to visually display the road surface conditions. An interactive visualization interface can also be made. Users can view the flatness details of different road sections by zooming in and rotating the model, and click on specific road sections to obtain detailed data and analysis reports.

[0150] Adaptive Adjustment Module:

[0151] 1. Basis and Frequency for Dynamic Adjustment of Parameters and Settings:

[0152] Based on the statistical analysis results of real-time data, evaluate and adjust the parameters in data standardization, association strategies, data fusion algorithms, and real-time update and feedback mechanisms once an hour. For example, if the data abnormality rate of a certain detection device exceeds 20% within 3 consecutive hours, automatically reduce the weight of this device in data fusion, and at the same time check whether the device needs maintenance. If it is found that the change trend of the road surface flatness of a certain road section deviates from the prediction result of the association model by more than 15%, adjust the relevant parameters in the association strategy, such as increasing or decreasing the association weight related to traffic flow, and retrain the machine learning model to adapt to the new changes.

[0153] 2. Continuous Improvement Measures for Optimizing the Data Processing Flow:

[0154] Regularly (e.g., monthly) review and evaluate the entire data processing process. Collect user feedback, including evaluations from road maintenance personnel on data accuracy and the practicality of processing results. Based on this feedback and evaluation results, comprehensively optimize the system, such as updating data fusion algorithms, improving the display effect of the visualization module, etc., to continuously enhance the flexibility and adaptability of data processing and the performance of the entire system.

[0155] Embodiment 3: Obviously, this data processing method can also be applied to: in a mountain highway road test signal detection system: data processing for pavement skid resistance performance detection, pavement damage condition detection, subgrade settlement detection, subgrade humidity detection, bridge structure detection, traffic sign detection, etc. Details are not elaborated herein.

Claims

1. A data processing method for a mountain highway road test signal detection system, characterized in that: The following steps are involved: S1: Data standardization: Standardize the raw data obtained from the mountain highway road test signal detection system to ensure that data from different sources and in different formats can be uniformly expressed and compared; S2: Data association strategy: Use predetermined association rules and methods to conduct association analysis on the standardized data and establish logical relationships between the data to reveal the intrinsic connection between road conditions and detection signals; S3: Data fusion algorithm selection and application: According to data characteristics and analysis requirements, select appropriate data fusion algorithms to fuse the associated multi-source data to generate more accurate and comprehensive road condition assessment information; S4: Real-time update and feedback mechanism: Establish a real-time data update system to ensure the timeliness and accuracy of the detection data, and through the feedback mechanism, promptly transmit the processing results to the relevant road maintenance and management departments so that timely response measures can be taken; It also includes: establishing a unified data management platform for integrating and collaboratively processing all data involved in steps S1 to S4 to achieve a comprehensive assessment of the road conditions of mountain highways.

2. The data processing method of a mountain highway road test signal detection system according to claim 1, characterized in that: The S1 also includes identifying and processing abnormal data to improve data quality.

3. The data processing method of a mountain highway road test signal detection system according to claim 1, characterized in that: In S2, an association method based on timestamp, geographic location or signal characteristics is adopted to achieve accurate matching and association of data.

4. The data processing method of a mountain highway road test signal detection system as claimed in claim 1, characterized in that: The S2 also includes a data association model based on a machine learning or deep learning algorithm, which can automatically identify and learn potential association patterns between data, further improving the accuracy and efficiency of data association analysis.

5. The data processing method of a mountain highway road test signal detection system as claimed in claim 1, characterized in that: In S3, according to the characteristics of the data and the analysis requirements, one or more data fusion algorithms among weighted average method, Kalman filter method, Bayesian network method or neural network method are selected to process the data.

6. The data processing method of a mountain highway road test signal detection system as claimed in claim 1, characterized in that: S4 also includes establishing a data quality monitoring system to monitor the data update and feedback process in real time to ensure the accuracy and timeliness of the data.

7. The data processing method of a mountain highway road test signal detection system as claimed in claim 1, characterized in that: The data management platform in S1 also includes a data visualization module for intuitively displaying the processed data in the form of charts, reports, etc., to facilitate user understanding and analysis.

8. The data processing method of a mountain highway road test signal detection system as claimed in claim 1, characterized in that: The S1 also includes an adaptive adjustment module, which can dynamically adjust the parameters and settings in data standardization, association strategy, data fusion algorithm, and real-time update and feedback mechanism according to the actual operation status and data changes of the road test signal detection system, so as to optimize the data processing flow and improve the flexibility and adaptability of data processing.

Images