Road infrastructure health monitoring and evaluation method and device and medium

Abstract

The invention relates to the technical field of data processing, and discloses a road infrastructure health monitoring and evaluation method and device, equipment and a medium. The method comprises the following steps: performing multi-mode acquisition and multi-path transmission processing on data acquired by an intelligent sensor network to obtain an original monitoring data set; cleaning, standardizing and fusing the original monitoring data set to obtain preprocessed structured data; performing multi-dimensional spatial correlation analysis on the structured data through a multi-scale time sequence analysis algorithm to obtain health state evaluation data of the road infrastructure; based on the health state evaluation data, performing dynamic threshold calculation and intelligent diagnosis on the road infrastructure to obtain early warning information and a preliminary diagnosis result; and performing expert knowledge base matching and multi-criterion decision analysis on the early warning information and the preliminary diagnosis result to obtain a maintenance decision suggestion and an optimization scheme. The road infrastructure health monitoring and evaluation efficiency and accuracy are improved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a method, device and medium for monitoring and evaluating the health of highway infrastructure. Background Art

[0002] As the scale of highway infrastructure continues to expand and its service life increases, its health status monitoring and evaluation becomes extremely important. Traditional highway infrastructure monitoring methods mainly rely on manual regular inspections and static testing, which have problems such as low monitoring frequency, limited coverage, and untimely data acquisition. In recent years, with the development of the Internet of Things, big data, and artificial intelligence technologies, intelligent sensor networks and data analysis technologies have begun to be applied to the health monitoring of highway infrastructure. These technologies can achieve real-time, continuous, and comprehensive monitoring of highway infrastructure, providing a more scientific basis for infrastructure maintenance and management.

[0003] However, the existing highway infrastructure health monitoring system still has some shortcomings. First, the deployment of sensor networks lacks systematicity and adaptability, and it is difficult to optimize the configuration for different types of infrastructure. Second, the energy consumption and security issues in the data collection and transmission process have not been fully addressed. Third, the processing and analysis capabilities of massive multi-source heterogeneous data are limited, making it difficult to fully mine the valuable information contained in the data. Finally, the existing system needs to be improved in terms of early warning mechanism, diagnostic accuracy and maintenance decision support, making it difficult to achieve accurate assessment and timely maintenance of the health status of highway infrastructure. Summary of the invention

[0004] The present application provides a method, device and medium for monitoring and evaluating the health of highway infrastructure, which are used to improve the efficiency and accuracy of monitoring and evaluating the health of highway infrastructure.

[0005] In a first aspect, the present application provides a method for monitoring and assessing the health of highway infrastructure, which comprises: deploying multiple types of sensors at key points of highway infrastructure to obtain an intelligent sensor network; performing multi-mode acquisition and multi-path transmission processing on data collected by the intelligent sensor network to obtain an original monitoring data set; cleaning, standardizing and fusing the original monitoring data set to obtain pre-processed structured data; performing multi-dimensional spatial correlation analysis on the structured data through a multi-scale time series analysis algorithm to obtain health status assessment data of the highway infrastructure; performing dynamic threshold calculation and intelligent diagnosis on the highway infrastructure based on the health status assessment data to obtain early warning information and preliminary diagnosis results; performing expert knowledge base matching and multi-criteria decision analysis on the early warning information and preliminary diagnosis results to obtain maintenance decision recommendations and optimization plans.

[0006] In a second aspect, the present application provides a highway infrastructure health monitoring and assessment device, the highway infrastructure health monitoring and assessment device comprising: A deployment module is used to deploy multiple types of sensors at key points of highway infrastructure to obtain a smart sensor network; A processing module, used for performing multi-mode collection and multi-path transmission processing on the data collected by the intelligent sensor network to obtain an original monitoring data set; A fusion module is used to clean, standardize and fuse the original monitoring data set to obtain pre-processed structured data; An analysis module, used to perform multi-dimensional spatial correlation analysis on the structured data by using a multi-scale time series analysis algorithm to obtain health status assessment data of highway infrastructure; A diagnosis module, used to perform dynamic threshold calculation and intelligent diagnosis on highway infrastructure based on the health status assessment data, and obtain early warning information and preliminary diagnosis results; The decision-making module is used to perform expert knowledge base matching and multi-criteria decision analysis on the warning information and preliminary diagnosis results to obtain maintenance decision suggestions and optimization plans.

[0007] The third aspect of the present application provides a computer device, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor and the memory communicate via a bus, and when the machine-readable instructions are executed by the processor, the steps of the above-mentioned highway infrastructure health monitoring and assessment method are performed.

[0008] A fourth aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned highway infrastructure health monitoring and assessment method.

[0009] In the technical solution provided by this application, by deploying multiple types of sensors at key points of highway infrastructure to form an intelligent sensor network, comprehensive and accurate monitoring of infrastructure is achieved, and the coverage and accuracy of data collection are improved. Secondly, the use of multi-mode acquisition and multi-path transmission processing methods not only improves the flexibility and adaptability of data acquisition, but also enhances the reliability and efficiency of data transmission, effectively solving the problems of untimely data acquisition and unstable transmission in traditional methods. The original monitoring data set is cleaned, standardized and fused, which significantly improves the data quality and lays a solid foundation for subsequent analysis. Through the multi-scale time series analysis algorithm, multi-dimensional spatial correlation analysis of structured data is carried out, and the implicit spatiotemporal patterns and associations in the data are deeply excavated, which greatly improves the accuracy and comprehensiveness of health status assessment. Based on the health status assessment data, dynamic threshold calculation and intelligent diagnosis are carried out to achieve timely warning and accurate diagnosis of potential risks of highway infrastructure, and greatly improve the sensitivity and reliability of the early warning system. Finally, by matching the early warning information and preliminary diagnosis results with expert knowledge bases and multi-criteria decision analysis, scientific and reasonable maintenance decision suggestions and optimization plans are obtained, which not only improves the accuracy and operability of decision-making, but also realizes the optimal allocation of maintenance resources. Overall, this method builds a complete closed loop from data collection, transmission, processing, analysis to decision support, realizing real-time monitoring, accurate assessment and intelligent maintenance of the health status of highway infrastructure, and significantly improving the management efficiency and safety of highway infrastructure. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0011] Figure 1 This is a schematic diagram of an embodiment of a method for monitoring and assessing the health of highway infrastructure in an embodiment of the present application; Figure 2 This is a schematic diagram of an embodiment of a highway infrastructure health monitoring and assessment device in an embodiment of the present application; Figure 3 It is a schematic diagram of the structure of a computer device in an embodiment of the present application. DETAILED DESCRIPTION

[0012] Embodiments of the present application provide a method, device and medium for monitoring and evaluating the health of highway infrastructure. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0013] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 , an embodiment of the highway infrastructure health monitoring and assessment method in the embodiment of the present application includes: Step S101: deploy multiple types of sensors at key points of highway infrastructure to obtain an intelligent sensor network; Step S102, performing multi-mode collection and multi-path transmission processing on the data collected by the intelligent sensor network to obtain an original monitoring data set; Step S103: cleaning, standardizing and fusing the original monitoring data set to obtain pre-processed structured data; Step S104: Perform multi-dimensional spatial correlation analysis on the structured data using a multi-scale time series analysis algorithm to obtain health status assessment data of highway infrastructure; Step S105: Based on the health status assessment data, dynamic threshold calculation and intelligent diagnosis are performed on the highway infrastructure to obtain early warning information and preliminary diagnosis results; Step S106: Perform expert knowledge base matching and multi-criteria decision analysis on the warning information and the preliminary diagnosis results to obtain maintenance decision suggestions and optimization solutions.

[0014] It is understandable that the execution subject of the present application may be a highway infrastructure health monitoring and assessment device, or a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0015] Specifically, multiple types of sensors are deployed at key points of highway infrastructure to form an intelligent sensor network. This involves a detailed analysis of the highway structure to determine the locations that need to be monitored, such as bridge supports and areas prone to pavement cracks. According to different monitoring needs, appropriate types of sensors are selected and installed, such as strain sensors, displacement sensors, vibration sensors, etc. These sensors form a network through wireless communication technology to achieve real-time data collection and transmission. Subsequently, the data collected by the intelligent sensor network is processed by multi-mode acquisition and multi-path transmission to obtain the original monitoring data set. Multi-mode acquisition includes timed acquisition, triggered acquisition, and on-demand acquisition to adapt to the characteristics of different monitoring parameters. For example, for slowly changing parameters such as pavement temperature, low-frequency timed acquisition can be used; while for rapidly changing parameters such as vehicle load, high-frequency triggered acquisition is used. Multi-path transmission uses mesh network technology to ensure that data can be transmitted reliably even if some sensors or communication links fail.

[0016] Next, the original monitoring data set is cleaned, standardized and fused to obtain pre-processed structured data. During the data cleaning process, statistical methods are used to identify and process outliers and missing values. Standardization unifies different types of data to the same scale for subsequent analysis. Data fusion comprehensively considers the relevance and complementarity of multi-source data to improve the reliability and integrity of the data. Then, the multi-dimensional spatial correlation analysis of the structured data is performed through the multi-scale time series analysis algorithm to obtain the health status assessment data of the highway infrastructure. Using technologies such as wavelet transform, the change characteristics of the data are analyzed at different time scales, and the spatial correlation is considered to comprehensively assess the health status of the infrastructure. Based on the health status assessment data, dynamic threshold calculation and intelligent diagnosis are performed on the highway infrastructure to obtain early warning information and preliminary diagnosis results. The dynamic threshold calculation takes into account the influence of environmental factors, usage intensity, etc., making the early warning more accurate. Intelligent diagnosis uses machine learning algorithms to identify and classify potential problems based on historical data and expert experience. Finally, expert knowledge base matching and multi-criteria decision analysis are performed on the early warning information and preliminary diagnosis results to obtain maintenance decision recommendations and optimization plans. Combined with historical cases and solutions in the expert knowledge base, and taking into account multiple criteria such as safety, economy, and feasibility, the optimal maintenance decision-making recommendations are given.

[0017] For example, in the health monitoring process of a highway bridge, strain sensors, acceleration sensors, etc. are first installed on key components of the bridge, such as the main beam and piers. These sensors collect data every 10 minutes, and high-frequency collection is triggered when a large vehicle passes. After the raw data is cleaned, outliers caused by sensor failure are removed, and some missing data are supplemented by interpolation. After standardization, different types of data are unified into the range of 0-1. Multi-scale time series analysis found that the strain data of the bridge main beam showed periodic changes on the daily scale, which was related to traffic flow; there was a slow growth trend on the monthly scale, which may indicate structural fatigue. Spatial correlation analysis found that the strain changes near the piers were large. The dynamic threshold calculation took into account the impact of seasonal temperature changes and appropriately increased the warning threshold in summer. Based on this information and combined with historical data, the intelligent diagnosis system preliminarily diagnosed that the bridge bearings may be slightly degraded. Finally, the system gave maintenance suggestions to increase the frequency of bearing inspections and consider replacing bearings in the near future.

[0018] In the embodiment of the present application, by deploying multiple types of sensors at key points of highway infrastructure to form an intelligent sensor network, comprehensive and accurate monitoring of infrastructure is achieved, and the coverage and accuracy of data collection are improved. Secondly, the multi-mode acquisition and multi-path transmission processing method is adopted, which not only improves the flexibility and adaptability of data acquisition, but also enhances the reliability and efficiency of data transmission, and effectively solves the problems of untimely data acquisition and unstable transmission in traditional methods. The original monitoring data set is cleaned, standardized and fused, which significantly improves the data quality and lays a solid foundation for subsequent analysis. The multi-dimensional spatial correlation analysis of structured data is carried out through the multi-scale time series analysis algorithm, and the implicit spatiotemporal patterns and associations in the data are deeply excavated, which greatly improves the accuracy and comprehensiveness of health status assessment. Based on the health status assessment data, dynamic threshold calculation and intelligent diagnosis are carried out to achieve timely warning and accurate diagnosis of potential risks of highway infrastructure, and greatly improve the sensitivity and reliability of the early warning system. Finally, by matching the early warning information and preliminary diagnosis results with expert knowledge base and multi-criteria decision analysis, scientific and reasonable maintenance decision suggestions and optimization schemes are obtained, which not only improves the accuracy and operability of decision-making, but also realizes the optimal allocation of maintenance resources. Overall, this method builds a complete closed loop from data collection, transmission, processing, analysis to decision support, realizing real-time monitoring, accurate assessment and intelligent maintenance of the health status of highway infrastructure, and significantly improving the management efficiency and safety of highway infrastructure.

[0019] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) Analyze the structural characteristics of highway infrastructure to obtain a distribution map of key monitoring points, match the sensor types on the distribution map of key monitoring points, and obtain a sensor deployment plan; (2) Perform energy consumption optimization calculation on the sensor deployment scheme to obtain a low-power sensor layout, and perform anti-interference evaluation on the low-power sensor layout to obtain a candidate sensor network topology; (3) Analyze the communication bandwidth of the candidate sensor network topology to obtain the data transmission path planning, and perform redundant design on the data transmission path planning to obtain the fault-tolerant sensor network structure; (4) Design a hierarchical architecture for the fault-tolerant sensor network structure to obtain a multi-level sensor network architecture, and configure edge computing capabilities for the multi-level sensor network architecture to obtain an initial intelligent sensor network; (5) The initial intelligent sensor network is dynamically and adaptively adjusted through a self-organizing algorithm to obtain a self-optimizing sensor network, and the self-optimizing sensor network is securely encrypted to obtain an intelligent sensor network.

[0020] Specifically, the structural characteristics of highway infrastructure are analyzed to obtain a distribution map of key monitoring points. This involves a comprehensive assessment of the structural type, material properties, load conditions, etc. of the highway to identify vulnerable parts and key structural points. For example, for bridge structures, key monitoring points include locations such as the mid-span of the main beam, supports, and bridge piers; for pavement structures, they include areas prone to cracks and settlement. Based on these key monitoring points, sensor type matching is performed to select suitable sensor types, such as strain sensors, displacement sensors, vibration sensors, etc., to obtain a preliminary sensor deployment plan. Next, the sensor deployment plan is energy-optimized and calculated to obtain a low-power sensor layout. Considering factors such as the power consumption characteristics of the sensor, data acquisition frequency, and communication distance, optimization algorithms such as genetic algorithms or particle swarm optimization algorithms are used to obtain a layout plan with the lowest energy consumption while meeting monitoring requirements. Subsequently, the low-power sensor layout is evaluated for anti-interference performance, considering factors such as environmental noise and electromagnetic interference to ensure the accuracy and reliability of sensor data, and finally obtain a candidate sensor network topology. Communication bandwidth analysis of the candidate sensor network topology is the next key step, which aims to ensure that the network can efficiently transmit a large amount of monitoring data. The optimal data transmission path planning is obtained by calculating the data generation rate of each node, network topology, communication protocol characteristics, etc. In order to improve the reliability of the network, it is also necessary to design the data transmission path planning redundantly and add backup paths to obtain a sensor network structure with fault tolerance.

[0021] Subsequently, the fault-tolerant sensor network structure is designed in a hierarchical architecture to obtain a multi-level sensor network architecture. This architecture usually includes a perception layer, a gateway layer, and an application layer. The perception layer is responsible for data collection, the gateway layer is responsible for data aggregation and preliminary processing, and the application layer is responsible for data analysis and decision support. On this basis, the edge computing capability is configured for the multi-level sensor network architecture, and edge computing nodes are deployed at the gateway layer to realize local processing and screening of data, reduce the amount of data transmitted to the central server, and improve the real-time performance and efficiency of the system, thereby obtaining an initial intelligent sensor network. Finally, the initial intelligent sensor network is dynamically and adaptively adjusted through a self-organizing algorithm to obtain a self-optimizing sensor network. The self-organizing algorithm can automatically adjust the network topology, routing strategy, and data acquisition parameters according to factors such as network status and environmental changes, thereby improving the adaptability and efficiency of the network. The self-optimizing sensor network is securely encrypted, and encryption algorithms such as AES and RSA are used to ensure the security of data transmission and storage, and finally form a complete intelligent sensor network.

[0022] For example, during the deployment of a health monitoring system for a highway bridge, 20 key monitoring points were first determined through structural analysis, including 5 in the span of the main beam, 8 in the support, 4 in the pier and 3 in the approach bridge. According to the monitoring requirements, 15 strain sensors, 10 acceleration sensors and 5 displacement sensors were selected. Through energy consumption optimization calculation, the optimal sampling frequency and transmission power of the sensors were determined, reducing the total network power consumption from the original 100W to 60W. The anti-interference assessment found that there were high-voltage lines under the bridge, and the positions of some sensors were fine-tuned. The communication bandwidth analysis showed that the data transmission volume during peak hours could reach 10MB / s. Based on this, a three-layer network structure was designed, including 30 sensor nodes, 5 gateway nodes and 1 central server. Edge computing capabilities were configured at the gateway nodes, which can compress and preliminarily analyze the raw data, reducing the amount of data transmitted to the central server by 40%. The self-organizing algorithm enables the network to automatically adjust the sampling frequency according to the traffic flow, increase the sampling frequency when the traffic volume is large, and reduce the sampling frequency when the traffic volume is small, which not only ensures data quality but also saves energy. Finally, the entire network uses the AES-256 encryption algorithm to protect data transmission security, forming a comprehensive, efficient and secure smart sensor network.

[0023] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) Configure the collection frequency of the intelligent sensor network to obtain a multi-mode data collection strategy, and set a threshold trigger mechanism for the multi-mode data collection strategy to obtain an adaptive collection scheme; (2) Performing data compression processing on the adaptive acquisition scheme to obtain a compressed data stream, and encrypting and encoding the compressed data stream to obtain a secure data packet; (3) Perform network topology analysis on the security data packets to obtain a multi-path transmission scheme, and perform load balancing calculation on the multi-path transmission scheme to obtain an optimized transmission path; (4) Select the transmission protocol for the optimized transmission path to obtain the hybrid network transmission strategy, and conduct reliability evaluation on the hybrid network transmission strategy to obtain the fault-tolerant transmission mechanism; (5) The fault-tolerant transmission mechanism is adjusted in real time through a dynamic routing algorithm to obtain an adaptive transmission network. The data transmitted by the adaptive transmission network is decrypted and decompressed to obtain the original monitoring data set.

[0024] Specifically, the intelligent sensor network is configured with acquisition frequency to obtain a multi-mode data acquisition strategy. Different acquisition frequencies are set for each type of sensor according to the characteristics and importance of different monitoring parameters. For example, for environmental parameters that change slowly, such as temperature and humidity, a lower acquisition frequency is used; for structural response parameters, such as strain and vibration, a higher acquisition frequency is used. At the same time, a threshold trigger mechanism is set for the multi-mode data acquisition strategy. When the monitoring parameter exceeds the preset threshold, the acquisition frequency is automatically increased, thereby obtaining an adaptive acquisition scheme. This scheme can effectively reduce energy consumption and data storage pressure while ensuring data quality. Subsequently, the adaptive acquisition scheme is subjected to data compression processing to obtain a compressed data stream. Data compression uses a lossless compression algorithm, such as Huffman coding or LZW algorithm, to ensure data integrity. The compression process takes into account the characteristics of different types of data, and uses pre-processing methods such as differential coding for data with large fluctuations to further improve compression efficiency. The compressed data stream is then encrypted and encoded to obtain a secure data packet. The encryption process uses high-intensity encryption algorithms such as AES-256 to ensure the security of data transmission.

[0025] Next, the network topology analysis of the security data packet is performed to obtain a multi-path transmission scheme. Taking into account the location, connection relationship and communication capacity of each node in the network, multiple optional data transmission paths are designed. Then, the multi-path transmission scheme is load-balanced to obtain the optimized transmission path. The load-balancing algorithm takes into account factors such as bandwidth, delay, and energy consumption of each path, dynamically allocates data flows, and avoids the situation where some paths are overloaded while other paths are idle. The transmission protocol is selected for the optimized transmission path to obtain a hybrid network transmission strategy. According to the characteristics of different paths and the requirements of the transmitted data, an appropriate transmission protocol is selected. For example, for data with high real-time requirements, the UDP protocol can be used; for data with high reliability requirements, the TCP protocol is used. The mixed use of multiple protocols can ensure transmission efficiency while taking into account data reliability. Subsequently, the reliability of the hybrid network transmission strategy is evaluated, and the fault tolerance of the network is tested by simulating different failure scenarios, and finally a fault-tolerant transmission mechanism is obtained.

[0026] Finally, the fault-tolerant transmission mechanism is adjusted in real time through the dynamic routing algorithm to obtain an adaptive transmission network. The dynamic routing algorithm can adjust the routing strategy in time according to the real-time changes in the network status, such as node failures, link congestion, etc., to ensure the continuity and efficiency of data transmission. After the data arrives at the destination, the data transmitted by the adaptive transmission network is decrypted and decompressed to restore the original monitoring data set, providing a basis for subsequent data analysis and decision support.

[0027] For example, in a highway bridge health monitoring system, 100 sensor nodes are deployed, including temperature sensors, strain sensors, and acceleration sensors. The basic acquisition frequency of the temperature sensor is set to once per hour, the strain sensor is set to once per minute, and the acceleration sensor is set to 10 times per second. At the same time, a threshold trigger mechanism is set: when the temperature change exceeds 5℃ / hour, the temperature sensor acquisition frequency is increased to once every 10 minutes; when the strain exceeds 80% of the preset value, the strain sensor acquisition frequency is increased to once per second; when the acceleration exceeds 0.5g, the acceleration sensor acquisition frequency is increased to 100 times per second. The data compression adopts the LZW algorithm. For temperature data, the compression ratio reaches 10:1; for strain data, the compression ratio is 5:1; for acceleration data, the compression ratio is 3:1. The compressed data is encrypted using the AES-256 algorithm. After the network topology analysis, 3 main transmission paths and 2 backup paths are designed. The load balancing algorithm allocates 70% of the data flow to the main path and 30% to the backup path according to the real-time network status.

[0028] In terms of transmission protocol selection, UDP protocol is used for temperature data, and TCP protocol is used for strain and acceleration data. Reliability evaluation shows that the network can still maintain a data transmission success rate of more than 95% when simulating 20% ​​node failures. When the dynamic routing algorithm detects congestion on a main path, it can complete the route adjustment within 0.5 seconds and transfer the data flow to the backup path. Through this multi-level, adaptive data collection and transmission solution, the bridge health monitoring system successfully collected and transmitted more than 100GB of raw monitoring data during a one-month operation period, with a data loss rate of less than 0.1%.

[0029] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) Perform outlier detection on the original monitoring data set to obtain preliminary cleaned data, and perform missing value interpolation on the preliminary cleaned data to obtain a complete data set; (2) Perform time series alignment on the complete data set to obtain a synchronized data sequence, and perform noise filtering on the synchronized data sequence to obtain a smoothed data set; (3) Perform data type unification processing on the smoothed data set to obtain data in a unified format, and perform dimension normalization processing on the data in the unified format to obtain a standardized data set; (4) Perform multi-source data association analysis on the standardized data set to obtain a data association matrix, and perform feature extraction on the data association matrix to obtain a feature vector set; (5) The feature vector set is integrated through a multimodal data fusion algorithm to obtain a fused data structure, and the data quality of the fused data structure is evaluated to obtain preprocessed structured data.

[0030] Specifically, outlier detection is performed on the original monitoring data set. Statistical methods, such as the triple standard deviation method or the box plot method, are used to identify and mark abnormal data points. For the identified outliers, they are processed according to their nature: obviously erroneous data are deleted, while outliers that may be caused by special events are marked and retained to obtain preliminary cleaned data. Subsequently, the preliminary cleaned data are interpolated for missing values, and time series interpolation methods, such as linear interpolation or spline interpolation, are used to fill in data gaps to obtain a complete data set. Next, the complete data set is aligned in time series. Since the sampling frequencies of different sensors may be different, all data need to be adjusted to a unified time scale. This process involves data resampling, interpolating data with lower sampling frequencies and downsampling data with higher sampling frequencies, and finally obtaining a synchronized data sequence. Then, the synchronized data sequence is subjected to noise filtering, and methods such as sliding average, median filtering or wavelet transform are used to remove high-frequency noise to obtain a smoothed data set.

[0031] The next key step is to unify the data type of the smoothed data set. The data output by different sensors may have different formats and units. They need to be converted into a unified data type and unit system to obtain data in a unified format. Subsequently, the unified format data is dimensionally normalized. Common methods include minimum-maximum normalization or Z-score normalization, which maps data of different dimensions to the same scale to obtain a standardized data set. Multi-source data association analysis of standardized data sets is an important step in understanding the relationship between different monitoring parameters. This process uses correlation analysis methods, such as Pearson correlation coefficient or mutual information, to calculate the degree of association between different data sources and obtain a data association matrix. Then, the data association matrix is ​​subjected to feature extraction processing. Methods such as principal component analysis (PCA) or independent component analysis (ICA) can be used to extract the most representative features and obtain a set of feature vectors.

[0032] Finally, the feature vector set is integrated through a multimodal data fusion algorithm. Decision-level fusion or feature-level fusion methods, such as Dempster-Shafer evidence theory or Kalman filtering, can be used to integrate information from different sensors to obtain a fused data structure. The fused data structure is evaluated for data quality, including completeness, consistency, and accuracy, and finally the preprocessed structured data is obtained.

[0033] For example, strain sensors, acceleration sensors, and temperature sensors are deployed in a highway bridge health monitoring system. During the one-week monitoring period, the strain sensor collects data once a minute, the acceleration sensor collects data 10 times a second, and the temperature sensor collects data once an hour. Outlier detection found that 3 points in the strain data exceeded 3 times the standard deviation. After analysis, it was found that this was caused by the passage of large vehicles, so they were retained and marked. In the missing value interpolation process, it was found that 2 hours of temperature data were missing, and linear interpolation was used to fill it. In the time series alignment process, all data were adjusted to a frequency of one data point per minute. For acceleration data, the mean and maximum values ​​within each minute were calculated as the representative values ​​of the minute; for temperature data, the piecewise constant interpolation method was used to fill it to one data point per minute. Noise filtering uses a 5-point sliding average method to effectively remove high-frequency noise.

[0034] Data type unification processing converts strain data from microstrain units to dimensionless strain values, acceleration data from g to m / s², and temperature data from Fahrenheit to Celsius. Dimension normalization uses the minimum-maximum normalization method to map all data to the [0,1] interval. Multi-source data association analysis calculates the Pearson correlation coefficient between strain, acceleration, and temperature, and finds that the correlation coefficient between strain and acceleration is 0.78, indicating a strong positive correlation; the correlation coefficient between strain and temperature is 0.45, indicating a moderate positive correlation. The PCA method is used for feature extraction, and the first three principal components that can explain 90% of the variance are selected as feature vectors. Multimodal data fusion uses the Dempster-Shafer evidence theory, which comprehensively considers the reliability and importance of different sensor data. Data quality assessment shows that the integrity of the structured data after preprocessing reaches 99.9%, the consistency score is 0.95 (full score is 1), and the accuracy is within 2% by comparing with the manual sampling results.

[0035] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) Decompose the structured data into time series to obtain trend, seasonal and random components, and perform multi-scale wavelet transform on the trend, seasonal and random components to obtain multi-scale time features; (2) Perform autocorrelation analysis on multi-scale time features to obtain a time dependency graph, and perform periodic pattern recognition on the time dependency graph to obtain periodic feature data; (3) Perform spatial autocorrelation analysis on the periodic characteristic data to obtain the spatial correlation matrix, and perform geographically weighted regression analysis on the spatial correlation matrix to obtain spatially differentiated influencing factors; (4) Construct multi-dimensional indicators for spatial differentiation influencing factors to obtain a comprehensive evaluation index system, and conduct hierarchical analysis on the comprehensive evaluation index system to obtain an indicator weight allocation scheme; (5) The fuzzy comprehensive evaluation algorithm is used to integrate and analyze the indicator weight allocation scheme and multi-scale time characteristics to obtain a fuzzy evaluation matrix, which is then defuzzified to obtain highway infrastructure health status assessment data.

[0036] Specifically, the structured data is decomposed into time series, and the data is decomposed into trend, seasonal and random components using classic time series decomposition methods, such as X-12-ARIMA or STL (Seasonal and Trend decomposition using Loess) algorithms. The trend component reflects the long-term changes in the data, the seasonal component reflects the periodic changes, and the random component contains short-term fluctuations and noise. Subsequently, the three components are subjected to multi-scale wavelet transforms, usually discrete wavelet transform (DWT) or continuous wavelet transform (CWT) to capture the characteristics at different time scales. Wavelet transform can analyze signals simultaneously in the time and frequency domains and effectively extract multi-scale time features, which contain health status information of highway infrastructure at different time scales. Next, the multi-scale time features are subjected to autocorrelation analysis, the autocorrelation coefficients of different time lags are calculated, and the time dependency diagram is drawn. Autocorrelation analysis can reveal the intrinsic relationship of data in time and help identify periodic patterns and long-term dependencies. To identify periodic patterns in the time dependency graph, Fourier transform or periodogram analysis can be used to identify significant periodic features, such as daily, weekly or seasonal changes, to obtain periodic feature data. These periodic features reflect the impact of periodic factors (such as traffic flow, temperature changes, etc.) on highway infrastructure.

[0037] Subsequently, the spatial autocorrelation analysis of the periodic characteristic data was performed, and the global Moran's I or local Moran's I was calculated to obtain the spatial correlation matrix. Spatial autocorrelation analysis can reveal the spatial dependence between different monitoring points and help identify spatial clustering or abnormal areas. The spatial correlation matrix was subjected to geographically weighted regression analysis. This step takes into account the impact of spatial position on the variable relationship, can capture spatial non-stationarity, and obtain spatially differentiated influencing factors. These influencing factors reflect the spatial variation characteristics of the health status of highway infrastructure. Then, multi-dimensional indicators were constructed for the spatially differentiated influencing factors, and a comprehensive comprehensive evaluation index system was established by comprehensively considering aspects such as structural safety, functional applicability, and durability. This index system reflects all aspects of the health status of highway infrastructure. Then, a hierarchical analysis was performed on the comprehensive evaluation index system, and the analytic hierarchy process (AHP) was used to determine the relative importance of each indicator and obtain the indicator weight distribution scheme. The AHP method ensures the scientificity and rationality of weight distribution through expert scoring and consistency testing.

[0038] Finally, the indicator weight allocation scheme and multi-scale time characteristics are integrated and analyzed through the fuzzy comprehensive evaluation algorithm. Fuzzy comprehensive evaluation can handle the uncertainty and ambiguity in the evaluation process and is more in line with the actual situation. This step first establishes a membership function, converts quantitative indicators into fuzzy sets, and then combines the indicator weights to obtain a fuzzy evaluation matrix. The fuzzy evaluation matrix is ​​defuzzified, usually using the centroid method or the maximum membership method, and finally the highway infrastructure health status assessment data is obtained. This assessment data combines time series characteristics, spatial correlation and multi-dimensional indicators, and comprehensively reflects the health status of highway infrastructure.

[0039] For example, a highway bridge health monitoring system analyzes the strain data of the main beam. First, the daily average strain data of one year is decomposed into a time series to obtain the long-term growth trend (annual growth rate of 0.5%), seasonal fluctuations (high in summer and low in winter, with a fluctuation range of ±2%), and random components. The three components are subjected to a 5-level discrete wavelet transform to obtain time characteristics of different scales. Autocorrelation analysis found a strong 7-day periodicity (autocorrelation coefficient 0.82), reflecting the difference in traffic patterns on weekdays and weekends. Spatial autocorrelation analysis showed that the strain of the central span of the bridge was significantly positively correlated with the spans on both sides (Moran's I = 0.75). Geographically weighted regression analysis found that the strain growth rate of the bridge section close to the exit ramp was 20% higher than the average level. The comprehensive evaluation index system includes 10 indicators such as strain level, strain growth rate, and crack density. The weights are determined by the AHP method, where the weight of the strain level is 0.3 and the weight of the strain growth rate is 0.25. In the fuzzy comprehensive evaluation process, the strain level is divided into 5 levels according to the proportion of the design value, and the membership function is established. The final health status assessment result was 0.78 (out of a maximum score of 1), indicating that the overall condition of the bridge was good, but there was a slight deterioration trend. It was recommended to strengthen monitoring and formulate a preventive maintenance plan.

[0040] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) Compare and analyze the historical data of health status assessment data to obtain a trend chart, and identify abnormal patterns on the trend chart to obtain potential risk points; (2) Conduct environmental factor correlation analysis on potential risk points to obtain risk impact factors, and perform weight calculation on risk impact factors to obtain dynamic threshold adjustment parameters; (3) The dynamic threshold adjustment parameters are processed through the adaptive threshold algorithm to obtain the multi-level warning thresholds, and the sensitivity analysis of the multi-level warning thresholds is performed to obtain the warning trigger conditions; (4) Match the warning trigger conditions to the rule base to obtain a preliminary diagnosis rule set, and train the preliminary diagnosis rule set through a machine learning algorithm to obtain an intelligent diagnosis model; (5) Apply the intelligent diagnosis model to analyze the health status assessment data to obtain the probability distribution of fault types, and classify the probability distribution of fault types into risk levels to obtain early warning information and preliminary diagnosis results.

[0041] Specifically, the historical data is processed using time series analysis techniques, such as the moving average method or the exponential smoothing method, to generate a trend chart. The trend chart intuitively shows the long-term trend of changes in the health status of highway infrastructure. Subsequently, the trend chart is used for abnormal pattern recognition, and data points or intervals that are significantly different from the historical pattern are identified using algorithms such as the local anomaly factor (LOF) or the isolation forest algorithm. These abnormal points are marked as potential risk points. Next, the potential risk points are analyzed for environmental factor associations. This step considers the impact of environmental factors such as temperature, humidity, and traffic load on the health status of highway infrastructure. Through correlation analysis or principal component analysis, the environmental factors that have the most significant impact on the potential risk points are determined to obtain the risk influencing factors. Subsequently, the risk influencing factors are weighted, and the entropy weight method or the analytic hierarchy process (AHP) can be used to assign weights according to the contribution of each factor to the risk, and finally the dynamic threshold adjustment parameters are obtained. These parameters reflect the degree of influence of environmental factors on the warning threshold.

[0042] The dynamic threshold adjustment parameters are processed by the adaptive threshold algorithm to obtain a multi-level warning threshold. The adaptive threshold algorithm takes into account the historical data distribution, current environmental conditions and risk influencing factors to dynamically adjust the warning threshold. Common methods include threshold setting based on statistical distribution or fuzzy logic control. The multi-level warning threshold is usually divided into four levels: attention, warning, serious and emergency, and each level corresponds to a different degree of risk. Subsequently, a sensitivity analysis is performed on the multi-level warning threshold to evaluate the impact of threshold changes on the warning results, so as to obtain the optimal warning trigger condition. Matching the warning trigger condition with the rule base is a key step in intelligent diagnosis. The rule base contains diagnostic rules based on expert knowledge and historical case summaries, such as "if the crack width exceeds 2mm and the growth rate is greater than 0.1mm / month, it is judged as structural fatigue." By matching the warning trigger condition with the rule base, a preliminary diagnostic rule set is obtained. Then, the preliminary diagnostic rule set is trained by a machine learning algorithm to obtain an intelligent diagnosis model. Common machine learning algorithms include decision trees, random forests or support vector machines (SVM). These algorithms can learn from a large amount of historical data to improve the accuracy and generalization ability of diagnosis.

[0043] Finally, the intelligent diagnosis model is applied to the health status assessment data to analyze the probability distribution of fault types. The intelligent diagnosis model calculates the probability of various possible fault types based on the input health status assessment data. For example, the model may output that the probability of fatigue cracking in the bridge main beam is 60%, the probability of bearing damage is 30%, and the probability of other fault types is lower. Subsequently, the probability distribution of fault types is divided into risk levels. The risk matrix method is usually used to comprehensively consider the probability and severity of the fault, and the risk is divided into three levels: low, medium, and high. Finally, early warning information and preliminary diagnosis results are obtained.

[0044] For example, the health monitoring system of a highway bridge analyzes the strain data of the main beam. First, the daily average strain data of the past five years was exponentially smoothed to generate a trend chart. The local anomaly factor algorithm was used to identify three potential risk points, whose strain values ​​were more than 20% higher than the same period in history. The correlation analysis of environmental factors found that temperature and traffic load were the main factors affecting strain, with a correlation coefficient of 0.65 for temperature and 0.78 for traffic load. The entropy weight method was used to calculate the weight of temperature as 0.4 and the weight of traffic load as 0.6, which were used as dynamic threshold adjustment parameters. The adaptive threshold algorithm adjusted the original fixed warning threshold (such as 2000 microstrain) to a dynamic threshold based on these parameters. For example, in the summer when the temperature is high and the traffic is at peak, the threshold may be increased to 2200 microstrain; in the winter when the temperature is low and the traffic is at low, the threshold may be reduced to 1800 microstrain. Sensitivity analysis shows that a threshold change of ±5% will not significantly affect the accuracy of the warning. During the rule base matching process, it was found that the current situation met the rule of "the strain value exceeded the dynamic threshold for 30 consecutive days and the growth rate was greater than 5 microstrain / day", and it was initially judged that there might be structural fatigue. The intelligent diagnosis model was trained using the random forest algorithm. The model input included strain value, growth rate, temperature, traffic load and other characteristics, and the output was the probability of various fault types. The trained model was used to analyze the current data to obtain the probability distribution of fault types: structural fatigue 65%, support abnormality 20%, and other fault types with low probability. According to the risk matrix, structural fatigue was classified as a high risk level. Finally, the warning information was generated: "The main beam may have structural fatigue, the risk level is high, and detailed inspection and stress monitoring are recommended." This result comprehensively considers historical data, environmental factors, expert knowledge and machine learning algorithms, providing timely and accurate warning and diagnosis information for bridge management departments, which helps to take targeted maintenance measures.

[0045] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) Perform semantic analysis on the warning information and preliminary diagnosis results to obtain key diagnostic features, and search the expert knowledge base for the key diagnostic features to obtain a set of similar cases; (2) Calculate case similarity for similar case sets to obtain the best matching case, and extract solutions for the best matching case to obtain candidate maintenance strategies; (3) Construct a multi-criteria evaluation index for the candidate maintenance strategy to obtain a decision evaluation system, and perform a hierarchical analysis on the decision evaluation system to obtain a criterion weight vector; (4) Comprehensively analyze the criterion weight vector and candidate maintenance strategies through the fuzzy comprehensive evaluation method to obtain the strategy priority ranking, and then conduct a cost-benefit analysis on the strategy priority ranking to obtain the optimal maintenance decision; (5) Decompose the implementation steps of the optimal maintenance decision to obtain a detailed maintenance plan, and optimize the resource allocation of the detailed maintenance plan to obtain maintenance decision recommendations and optimization plans.

[0046] Specifically, semantic analysis is performed on the warning information and preliminary diagnosis results, and natural language processing techniques such as word frequency statistics and topic models are used to extract keywords and semantic features to obtain key diagnostic features. Subsequently, these key diagnostic features are searched in the expert knowledge base, and information retrieval techniques such as vector space models or latent semantic indexing are used to find historical cases similar to the current situation to form a set of similar cases. Next, the case similarity of the similar case set is calculated, usually using methods such as cosine similarity or Euclidean distance. The calculation formula is as follows: in, represents the similarity between cases A and B, and Respectively represent the values ​​of case A and B on the i-th feature, and n is the total number of features. Through this calculation, the best matching case is obtained, and the solution is extracted from it to form a candidate maintenance strategy. Multi-criteria evaluation indicators are constructed for the candidate maintenance strategy, including technical feasibility, economic benefits, environmental impact and other aspects, to establish a decision-making evaluation system. Subsequently, a hierarchical analysis is performed on the decision-making evaluation system, and the weight of each criterion is calculated using the analytic hierarchy process (AHP). The weight calculation formula of AHP is as follows: in, For the The weight of the criteria, is the element in the i-th row and j-th column of the judgment matrix, and n is the total number of criteria. Through this calculation, the criterion weight vector is obtained. Then, the criterion weight vector and the candidate maintenance strategy are comprehensively analyzed by the fuzzy comprehensive evaluation method. First, the fuzzy relationship matrix R is established, and then the fuzzy synthesis operation is performed with the weight vector W to obtain the comprehensive evaluation result B. The calculation formula is as follows: in, represents the fuzzy synthesis operation, m represents the number of evaluation levels, It represents the membership of the nth evaluation index to the mth evaluation level. Usually, the max-min operation or weighted average method is used. According to the comprehensive evaluation results, the strategy priority ranking is obtained. Subsequently, the strategy priority ranking is subjected to cost-benefit analysis, considering the implementation cost and expected benefit of each strategy, and the optimal maintenance decision is obtained.

[0047] Finally, the optimal maintenance decision is decomposed into implementation steps, and the overall decision is divided into specific operation steps to obtain a detailed maintenance plan. The detailed maintenance plan is optimized for resource allocation, considering the optimal allocation of resources such as manpower, equipment, and materials. Linear programming or heuristic algorithms are usually used to solve this optimization problem. Finally, maintenance decision suggestions and optimization plans are obtained. For example, the health monitoring system of a highway bridge detected fatigue cracks in the main beam, and the early warning information showed that "a 2mm wide crack appeared in the middle of the main beam, with a growth rate of 0.1mm / month, which is a high risk level." Semantic analysis extracts key diagnostic features: main beam, fatigue crack, 2mm width, and 0.1mm / month growth rate. Expert knowledge base search found 5 similar cases. The case similarity calculation uses the cosine similarity method, and the calculation results show that the similarity of the most similar case is 0.92. The solution to this case includes three candidate maintenance strategies: crack injection repair, carbon fiber reinforcement, and steel plate reinforcement. The multi-criteria evaluation indicators include four aspects: technical feasibility, economic cost, repair effect, and construction period impact. The weight vector W = (0.3, 0.25, 0.3, 0.15) was calculated by the AHP method. The scores of the three strategies obtained by fuzzy comprehensive evaluation were 0.82, 0.75, and 0.68 respectively. The cost-benefit analysis showed that the crack injection repair plan had the highest return on investment. The final maintenance decision recommendation was to adopt the crack injection repair plan. The detailed maintenance plan includes: traffic control, crack cleaning, injection equipment installation, injection operation, maintenance and other steps. The resource allocation optimization results show that 4 professional technicians and 2 injection equipment are required, with an estimated construction period of 3 days and a total cost of about 500,000 yuan. This plan minimizes the impact on traffic while ensuring the repair effect.

[0048] The above describes the highway infrastructure health monitoring and assessment method in the embodiment of the present application. The following describes the highway infrastructure health monitoring and assessment device in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the highway infrastructure health monitoring and assessment device includes: A deployment module 201 is used to deploy multiple types of sensors at key points of highway infrastructure to obtain an intelligent sensor network; The processing module 202 is used to perform multi-mode collection and multi-path transmission processing on the data collected by the smart sensor network to obtain an original monitoring data set; The fusion module 203 is used to clean, standardize and fuse the original monitoring data set to obtain pre-processed structured data; An analysis module 204 is used to perform multi-dimensional spatial correlation analysis on the structured data by using a multi-scale time series analysis algorithm to obtain health status assessment data of highway infrastructure; A diagnosis module 205, for performing dynamic threshold calculation and intelligent diagnosis on the highway infrastructure based on the health status assessment data, and obtaining early warning information and preliminary diagnosis results; The decision module 206 is used to perform expert knowledge base matching and multi-criteria decision analysis on the warning information and the preliminary diagnosis results to obtain maintenance decision suggestions and optimization solutions.

[0049] Through the collaboration of the above components, by deploying multiple types of sensors at key points of highway infrastructure, an intelligent sensor network is formed, which realizes comprehensive and accurate monitoring of infrastructure and improves the coverage and accuracy of data collection. Secondly, the adoption of multi-mode acquisition and multi-path transmission processing methods not only improves the flexibility and adaptability of data collection, but also enhances the reliability and efficiency of data transmission, effectively solving the problems of untimely data acquisition and unstable transmission in traditional methods. The original monitoring data set is cleaned, standardized and fused, which significantly improves the data quality and lays a solid foundation for subsequent analysis. The multi-dimensional spatial correlation analysis of structured data is carried out through the multi-scale time series analysis algorithm, and the implicit spatiotemporal patterns and associations in the data are deeply excavated, which greatly improves the accuracy and comprehensiveness of health status assessment. Based on the health status assessment data, dynamic threshold calculation and intelligent diagnosis are carried out to realize timely warning and accurate diagnosis of potential risks of highway infrastructure, greatly improving the sensitivity and reliability of the early warning system. Finally, by matching the early warning information and preliminary diagnosis results with expert knowledge base and multi-criteria decision analysis, scientific and reasonable maintenance decision suggestions and optimization plans are obtained, which not only improves the accuracy and operability of decision-making, but also realizes the optimal allocation of maintenance resources. Overall, this method builds a complete closed loop from data collection, transmission, processing, analysis to decision support, realizing real-time monitoring, accurate assessment and intelligent maintenance of the health status of highway infrastructure, and significantly improving the management efficiency and safety of highway infrastructure.

[0050] Based on the same technical concept, the embodiment of the present application also provides an electronic device. Figure 3 As shown, it is a schematic diagram of the structure of the electronic device 300 provided in the embodiment of the present application, including a processor 301, a memory 302, and a bus 303. Among them, the memory 302 is used to store execution instructions, including a memory 3021 and an external memory 3022; the memory 3021 here is also called an internal memory, which is used to temporarily store the operation data in the processor 301, and the data exchanged with the external memory 3022 such as a hard disk. The processor 301 exchanges data with the external memory 3022 through the memory 3021. When the electronic device 300 is running, the processor 301 communicates with the memory 302 through the bus 303.

[0051] The present application also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the highway infrastructure health monitoring and assessment method.

[0052] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0053] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0054] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A highway infrastructure health monitoring and assessment method, characterized in that: The highway infrastructure health monitoring and assessment method comprises: Deploy multiple types of sensors at key points of highway infrastructure to obtain a smart sensor network; Performing multi-mode collection and multi-path transmission processing on the data collected by the intelligent sensor network to obtain an original monitoring data set; Cleaning, standardizing and fusing the original monitoring data set to obtain preprocessed structured data; Performing multi-dimensional spatial correlation analysis on the structured data by using a multi-scale time series analysis algorithm to obtain health status assessment data of highway infrastructure; Based on the health status assessment data, dynamic threshold calculation and intelligent diagnosis are performed on the highway infrastructure to obtain early warning information and preliminary diagnosis results; Expert knowledge base matching and multi-criteria decision analysis are performed on the warning information and preliminary diagnosis results to obtain maintenance decision suggestions and optimization plans.

2. The highway infrastructure health monitoring and assessment method according to claim 1 is characterized in that: The above-mentioned deployment of multiple types of sensors at key points of highway infrastructure provides an intelligent sensor network, including: Analyze the structural characteristics of highway infrastructure to obtain a distribution map of key monitoring points, and match the sensor types on the distribution map of key monitoring points to obtain a sensor deployment plan; Performing energy consumption optimization calculation on the sensor deployment scheme to obtain a low-power sensor layout, and performing anti-interference evaluation on the low-power sensor layout to obtain a candidate sensor network topology; Performing communication bandwidth analysis on the candidate sensor network topology to obtain a data transmission path plan, and performing redundancy design on the data transmission path plan to obtain a fault-tolerant sensor network structure; Performing a hierarchical architecture design on the fault-tolerant sensor network structure to obtain a multi-level sensor network architecture, and configuring edge computing capabilities on the multi-level sensor network architecture to obtain an initial intelligent sensor network; The initial intelligent sensor network is dynamically and adaptively adjusted through a self-organizing algorithm to obtain a self-optimizing sensor network, and the self-optimizing sensor network is securely encrypted to obtain an intelligent sensor network.

3. The highway infrastructure health monitoring and assessment method according to claim 1 is characterized in that: The data collected by the intelligent sensor network is subjected to multi-mode collection and multi-path transmission processing to obtain an original monitoring data set, including: Performing collection frequency configuration on the intelligent sensor network to obtain a multi-mode data collection strategy, and setting a threshold trigger mechanism on the multi-mode data collection strategy to obtain an adaptive collection scheme; Performing data compression processing on the adaptive acquisition scheme to obtain a compressed data stream, and encrypting and encoding the compressed data stream to obtain a secure data packet; Performing network topology analysis on the security data packet to obtain a multi-path transmission scheme, and performing load balancing calculation on the multi-path transmission scheme to obtain an optimized transmission path; Selecting a transmission protocol for the optimized transmission path to obtain a hybrid network transmission strategy, and performing reliability evaluation on the hybrid network transmission strategy to obtain a fault-tolerant transmission mechanism; The fault-tolerant transmission mechanism is adjusted in real time through a dynamic routing algorithm to obtain an adaptive transmission network, and the data transmitted by the adaptive transmission network is decrypted and decompressed to obtain an original monitoring data set.

4. The highway infrastructure health monitoring and assessment method according to claim 1, characterized in that: The cleaning, standardization and fusion processing of the original monitoring data set to obtain pre-processed structured data includes: Performing outlier detection on the original monitoring data set to obtain preliminary cleaned data, and performing missing value interpolation processing on the preliminary cleaned data to obtain a complete data set; Performing time series alignment on the complete data set to obtain a synchronized data sequence, and performing noise filtering on the synchronized data sequence to obtain a smoothed data set; Performing data type unification processing on the smoothed data set to obtain unified format data, and performing dimension normalization processing on the unified format data to obtain a standardized data set; Performing multi-source data association analysis on the standardized data set to obtain a data association matrix, and performing feature extraction processing on the data association matrix to obtain a feature vector set; The feature vector set is integrated and processed by a multimodal data fusion algorithm to obtain a fused data structure, and data quality assessment is performed on the fused data structure to obtain pre-processed structured data.

5. The highway infrastructure health monitoring and assessment method according to claim 1, characterized in that: The multi-dimensional spatial correlation analysis of the structured data is performed by a multi-scale time series analysis algorithm to obtain highway infrastructure health status assessment data, including: Performing time series decomposition on the structured data to obtain trend, seasonality and random components, and performing multi-scale wavelet transform on the trend, seasonality and random components to obtain multi-scale time features; Performing autocorrelation analysis on the multi-scale time features to obtain a time dependency graph, and performing periodic pattern recognition on the time dependency graph to obtain periodic feature data; Performing spatial autocorrelation analysis on the periodic characteristic data to obtain a spatial correlation matrix, and performing geographically weighted regression analysis on the spatial correlation matrix to obtain spatially differentiated influencing factors; Constructing multi-dimensional indicators for the spatial differentiation influencing factors to obtain a comprehensive evaluation indicator system, and performing hierarchical analysis on the comprehensive evaluation indicator system to obtain an indicator weight allocation scheme; The indicator weight distribution scheme and the multi-scale time characteristics are integrated and analyzed through a fuzzy comprehensive evaluation algorithm to obtain a fuzzy evaluation matrix, and the fuzzy evaluation matrix is ​​defuzzified to obtain highway infrastructure health status assessment data.

6. The highway infrastructure health monitoring and assessment method according to claim 1, characterized in that: Based on the health status assessment data, dynamic threshold calculation and intelligent diagnosis are performed on the highway infrastructure to obtain early warning information and preliminary diagnosis results, including: Performing historical data comparison analysis on the health status assessment data to obtain a change trend graph, and performing abnormal pattern recognition on the change trend graph to obtain potential risk points; Performing environmental factor correlation analysis on the potential risk points to obtain risk impact factors, and performing weight calculation on the risk impact factors to obtain dynamic threshold adjustment parameters; The dynamic threshold adjustment parameter is processed by an adaptive threshold algorithm to obtain a multi-level warning threshold, and a sensitivity analysis is performed on the multi-level warning threshold to obtain a warning trigger condition; Matching the warning triggering conditions to a rule base to obtain a preliminary diagnosis rule set, and training the preliminary diagnosis rule set through a machine learning algorithm to obtain an intelligent diagnosis model; The intelligent diagnosis model is applied to analyze the health status assessment data to obtain a probability distribution of fault types, and the probability distribution of fault types is classified into risk levels to obtain early warning information and preliminary diagnosis results.

7. The highway infrastructure health monitoring and assessment method according to claim 1, characterized in that: The expert knowledge base matching and multi-criteria decision analysis are performed on the warning information and the preliminary diagnosis results to obtain maintenance decision suggestions and optimization plans, including: Performing semantic analysis on the warning information and preliminary diagnosis results to obtain key diagnostic features, and performing expert knowledge base retrieval on the key diagnostic features to obtain a similar case set; Calculating case similarity on the similar case set to obtain the best matching case, and extracting solutions from the best matching case to obtain candidate maintenance strategies; Constructing a multi-criteria evaluation index for the candidate maintenance strategy to obtain a decision evaluation system, and performing a hierarchical analysis on the decision evaluation system to obtain a criterion weight vector; Comprehensively analyzing the criterion weight vector and the candidate maintenance strategy through a fuzzy comprehensive evaluation method to obtain a strategy priority ranking, and performing a cost-benefit analysis on the strategy priority ranking to obtain an optimal maintenance decision; The optimal maintenance decision is decomposed into implementation steps to obtain a detailed maintenance plan, and resource allocation optimization is performed on the detailed maintenance plan to obtain maintenance decision suggestions and optimization plans.

8. A highway infrastructure health monitoring and assessment device, used to implement the highway infrastructure health monitoring and assessment method according to any one of claims 1 to 7, characterized in that: The highway infrastructure health monitoring and assessment device comprises: A deployment module is used to deploy multiple types of sensors at key points of highway infrastructure to obtain a smart sensor network; A processing module, used for performing multi-mode collection and multi-path transmission processing on the data collected by the intelligent sensor network to obtain an original monitoring data set; A fusion module is used to clean, standardize and fuse the original monitoring data set to obtain pre-processed structured data; An analysis module, used to perform multi-dimensional spatial correlation analysis on the structured data by using a multi-scale time series analysis algorithm to obtain health status assessment data of highway infrastructure; A diagnosis module, used to perform dynamic threshold calculation and intelligent diagnosis on highway infrastructure based on the health status assessment data, and obtain early warning information and preliminary diagnosis results; The decision-making module is used to perform expert knowledge base matching and multi-criteria decision analysis on the warning information and preliminary diagnosis results to obtain maintenance decision suggestions and optimization plans.

9. A computer device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the highway infrastructure health monitoring and assessment method as described in any one of claims 1 to 7 are performed.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the highway infrastructure health monitoring and assessment method according to any one of claims 1 to 7 is implemented.

Images