Traffic infrastructure safety monitoring method and system

Abstract

The invention discloses a traffic infrastructure safety monitoring method and system. The system comprises a data acquisition unit, a data transmission unit, a data analysis processing unit, a cloud network database, a storage unit and a visualization unit. The data acquisition unit is used for collecting traffic information; the data transmission unit is used for real-time transmission of data; the data analysis processing unit receives the monitoring data, performs data cleaning and preprocessing in combination with cloud network data, and then performs modeling analysis; and the cloud network database is used for providing traffic-related information published by the social media. The system is additionally provided with the cloud network database, and judgment and processing are carried out by providing traffic related information published by social media, utilizing traffic information data published by the social media and combining real-time data obtained at the data acquisition unit, so that the system can more effectively carry out pre-judgment preparation of traffic safety control; and thus, traffic safety control is facilitated.

Description

Technical Field

[0001] The present invention relates to the field of infrastructure monitoring, and in particular to a method and system for safety monitoring of transportation infrastructure. Background Art

[0002] Transportation infrastructure refers to various facilities and systems used to support transportation, including roads, bridges, tunnels, railways, airports, ports and public transportation systems. These infrastructures directly affect people's travel efficiency and quality of life, so it is very important to monitor the security of these infrastructures to ensure smooth and safe transportation.

[0003] The existing monitoring systems of this type first establish a data acquisition network through a variety of sensors, use the content collected by each sensor, after data processing, use model learning and analysis methods to analyze the data, and finally display the analyzed data through visualization tools to assist in judging the safety status of traffic facilities. Although the existing systems of this type can monitor the safety of traffic infrastructure in real time through the above-mentioned methods, and thus arrange and adjust the traffic conditions in real time, such system operation forms cannot cope with some emergencies, such as the usage and usage of infrastructure in some past time periods. Most of these phenomena are repeated in previous years, and are mostly temporary or emergencies. If such situations are not planned and controlled in advance, the safety of traffic will be affected. Therefore, a method and system for monitoring the safety of traffic infrastructure are proposed. Summary of the invention

[0004] The object of the present invention is to provide a method and system for monitoring the safety of traffic infrastructure to solve the problems raised in the above background technology.

[0005] To achieve the above-mentioned object, the present invention provides the following technical solutions: a traffic infrastructure safety monitoring system, comprising a data acquisition unit, a data transmission unit, a data analysis and processing unit, a cloud network database, a storage unit and a visualization unit;

[0006] The data acquisition unit forms a monitoring network through a variety of sensors and cameras, and is used to collect traffic information through the sensors and cameras;

[0007] The data transmission unit establishes a wireless communication network at each sensor and camera for real-time data transmission;

[0008] The data analysis and processing unit receives the monitoring data transmitted by the data transmission unit, cleans and preprocesses the data in combination with the cloud network data, and then performs modeling analysis on the cleaned and preprocessed data through modeling;

[0009] The cloud network database is used to provide traffic related information published on social media;

[0010] The storage unit is used for storing data;

[0011] The visualization unit displays the analysis results in a graphical form to assist in determining the safety status of traffic facilities.

[0012] Preferably, the types of sensors include environmental sensors, structural health monitoring sensors and vehicle monitors;

[0013] The environmental sensors include temperature and humidity sensors, rainfall sensors, wind speed sensors and infrared sensors, which are used to monitor meteorological conditions, including temperature, humidity and wind speed;

[0014] The structural health monitoring sensors include strain sensors, displacement sensors and piezoelectric sensors, which are used to monitor the structural status of transportation infrastructure, including strain, displacement and vibration;

[0015] The vehicle monitor uses a millimeter wave radar, which transmits electromagnetic waves and receives reflected signals to count vehicle flow, vehicle speed and vehicle type classification.

[0016] Preferably, the data analysis and processing unit includes a data deduplication module, a data filling module and a data standardization module, and data cleaning and preprocessing are performed through the deduplication module, the data filling module and the data standardization module.

[0017] Preferably, the deduplication module is used to ensure the uniqueness of each piece of data by removing redundant information;

[0018] The data filling module fills the missing data using interpolation and mean filling methods. The interpolation method estimates the missing data by using the relationship between existing data points, and the mean filling method replaces the missing value with the mean of other observations in the data set.

[0019] The data standardization module is used to unify the formats of data from different sources.

[0020] A method for monitoring the safety of traffic infrastructure, using the monitoring system, the monitoring method comprises the following steps:

[0021] S1, first establish a data acquisition unit and a data transmission unit, set up sensors and cameras for information acquisition in the area to be monitored, and after setting up the sensors and cameras, establish a wireless network communication for the transmission of the collected data;

[0022] S2, wireless network communication transmits the collected data to the data analysis and processing unit in real time, and encrypts the data during the transmission process;

[0023] S3, after receiving the data, the data analysis and processing unit first decrypts it, then cleans and pre-processes the data, and then retrieves the data from the cloud network database. Finally, it uses machine learning algorithms and statistical methods to combine the processed data with the data retrieved from the database for real-time analysis;

[0024] S4, storing the processed data in a cloud network database and performing data backup at the same time;

[0025] S5, visualizes real-time data and analysis results through dashboards and charts, and sets threshold areas for comparison.

[0026] Preferably, the setting type of the sensor is selected according to the data to be collected, and the time interval error of each sensor transmitting data is 0.5-0.8s, and the transmission time is calibrated every 5-7 days.

[0027] Preferably, the data is transmitted using end-to-end encryption, and the data is always kept encrypted during the entire process of being sent from the sensor to the receiving end.

[0028] Preferably, the data cleaning and preprocessing include data deduplication, data filling and data standardization.

[0029] Preferably, during the real-time analysis of the data, frequency domain features are first extracted through Fourier transform, or trend and periodic features are extracted through time series analysis. Then, based on whether the data is labeled, the corresponding algorithm is selected for learning, and then a trial model is used for training. Finally, the trained model is used to identify data points that are significantly different from the normal pattern.

[0030] Preferably, the selection of the algorithm includes a classification or regression algorithm and a clustering algorithm;

[0031] The classification or regression algorithm includes decision tree, random forest, support vector machine;

[0032] The clustering algorithms include K-means clustering algorithm, DBSCAN algorithm and convolutional neural network algorithm.

[0033] Technical effects and advantages of the present invention:

[0034] (1) The system of the present invention is equipped with a cloud network database, which provides traffic-related information published on social media, and uses the traffic information data published on social media in combination with the real-time data obtained from the data acquisition unit for judgment and processing, so that the system can more effectively make advance judgments and preparations for traffic safety control, thereby being more conducive to traffic safety control;

[0035] (2) The monitoring method of the present invention performs encryption and decryption processing during data transmission, so that the security of the data can be guaranteed during subsequent processing. At the same time, data is retrieved from the cloud network database, and traffic control and analysis are performed using multi-data combined judgment to ensure smooth and safe transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a block diagram of the monitoring system of the present invention.

[0037] Figure 2 It is an operation flow chart of the monitoring method of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] Embodiment 1, the present invention provides as follows Figure 1 A traffic infrastructure safety monitoring system shown includes a data acquisition unit, a data transmission unit, a data analysis and processing unit, a cloud network database, a storage unit and a visualization unit;

[0040] The data acquisition unit forms a monitoring network through a variety of sensors and cameras, which are used to collect traffic information;

[0041] Specifically, the types of sensors include environmental sensors, structural health monitoring sensors, and vehicle monitors;

[0042] Environmental sensors Temperature and humidity sensors, rain sensors, wind speed sensors and infrared sensors are used to monitor meteorological conditions, including temperature, humidity and wind speed;

[0043] It should be noted that environmental sensors include temperature and humidity sensors, rainfall sensors, wind speed sensors and infrared sensors, which are used for environmental monitoring to determine the environmental conditions of transportation infrastructure so that the system can obtain environmental data.

[0044] Structural health monitoring sensors include strain sensors, displacement sensors, and piezoelectric sensors, which are used to monitor the structural status of transportation infrastructure, including strain, displacement, and vibration;

[0045] It should be noted that strain sensors are used to measure the strain (deformation degree) of materials. Common ones include fiber optic strain sensors and resistance strain gauges. A specific example of a fiber optic strain sensor is a fiber grating sensor, which works by using the sensitivity of the fiber grating to external strain. This sensor includes components such as a tube shell, an end cover, a fiber grating, and a strain desensitizing beam. The fiber grating is pre-tensioned and fixed on the end cover. When it is strained, the periodic structure of the fiber grating changes, causing the wavelength of the reflected or transmitted light to change, so that the magnitude of the strain can be measured. The principle of the resistance strain gauge is to use the change in the resistance of the resistance material to sense the strain of the object. When the resistance strain gauge is installed on the surface of the component and subjected to a load, the slight deformation (elongation or shortening) of the component will cause the sensitive grid of the strain gauge to deform, thereby causing the resistance of the strain gauge to change. This resistance change can be measured by a resistance strain gauge and converted into a strain value, or an analog electrical signal (such as voltage or current) proportional to the strain can be output, and then recorded by a recorder, or the data can be processed using a computer to obtain the measured stress or strain data.

[0046] The displacement sensor uses a laser displacement sensor, which is a sensor that uses laser technology for measurement. It consists of a laser, a laser detector, and a measurement circuit. The laser sensor is a new type of measuring instrument. It can accurately measure the position, displacement, and other changes of the measured object without contact.

[0047] Piezoelectric sensor is a sensor based on piezoelectric effect. It is a self-generating and electromechanical conversion sensor. Its sensitive element is made of piezoelectric material. When the piezoelectric material is subjected to force, the surface generates charge. After the charge is amplified and the impedance is transformed by the charge amplifier and the measurement circuit, it becomes an electrical output proportional to the external force. Piezoelectric sensors are used to measure non-electrical physical quantities that can be converted into electricity. Its advantages are wide bandwidth, high sensitivity, high signal-to-noise ratio, simple structure, reliable operation and light weight. The disadvantage is that some piezoelectric materials require moisture-proof measures, and the output DC response is poor, and a high input impedance circuit or charge amplifier is needed to overcome this defect.

[0048] Vehicle monitors are used to count vehicle flow, speed and vehicle type classification.

[0049] The data transmission unit establishes a wireless communication network at each sensor and camera for real-time data transmission;

[0050] It should be noted that data transmission is carried out by wireless transmission, and encryption is performed during the wireless data transmission process to ensure the real-time performance of data transmission while ensuring the security of data transmission.

[0051] The data analysis and processing unit receives the monitoring data transmitted by the data transmission unit, and performs data cleaning and preprocessing in combination with the cloud network data, and then performs modeling analysis on the cleaned and preprocessed data through modeling;

[0052] Specifically, the data analysis and processing unit includes a data deduplication module, a data filling module and a data standardization module, and data cleaning and preprocessing are performed through the deduplication module, the data filling module and the data standardization module.

[0053] Furthermore, the deduplication module is used to ensure the uniqueness of each piece of data by removing redundant information;

[0054] The data imputation module handles missing values ​​by filling in missing data using interpolation and mean imputation methods;

[0055] The data standardization module is used to unify the format of data from different sources to facilitate subsequent analysis.

[0056] A cloud-based web database is used to provide traffic-related information posted on social media;

[0057] The storage unit is used for storing data;

[0058] It should be noted that the storage unit uses a data lake (such as Amazon S3, Azure Data Lake) to store raw data, which facilitates flexible access to data. Alternatively, a data warehouse is used for storage, and structured data is stored in a data warehouse (such as Amazon Redshift, Google BigQuery) to support efficient query and analysis.

[0059] The visualization unit displays the analysis results in a graphical form to assist in determining the safety status of traffic facilities.

[0060] It should be noted that the visualization unit can use Tableau and Power BI to display traffic trends.

[0061] Embodiment 2, the present invention provides as follows Figure 2 A traffic infrastructure safety monitoring method is shown, using the monitoring system of embodiment 1, and the monitoring method includes the following steps:

[0062] S1, first establish a data acquisition unit and a data transmission unit, set up sensors and cameras for information acquisition in the area to be monitored, and after setting up the sensors and cameras, establish a wireless network communication for the transmission of the collected data;

[0063] Specifically, the setting type of the sensor is selected according to the data to be collected, and the time interval error of each sensor transmitting data is 0.5-0.8s, and the transmission time is calibrated every 5-7 days.

[0064] It should be noted that each sensor uses a similar time interval and transmits data while trying to keep it synchronized as much as possible, so that the data transmission of each sensor is synchronized and stable. Since errors are inevitable in time, they are reset and calibrated every 5-7 days to avoid excessive error accumulation, making the data transmission time more synchronized and timely.

[0065] S2, wireless network communication transmits the collected data to the data analysis and processing unit in real time, and encrypts the data during the transmission process;

[0066] Specifically, data transmission uses end-to-end encryption, and the data remains encrypted throughout the entire process of being sent from the sensor to the receiving end.

[0067] It should be noted that end-to-end encryption is a security technology that ensures that data remains encrypted during transmission. In this way, only the sender and receiver of the data can decrypt and access the data, and any node in the middle (such as a server or network provider) cannot obtain the plaintext data. In the entire process of sending data from the sensor to the receiver, the key steps to implement end-to-end encryption include: data encryption, after the data is generated at the sensor end, the data is encrypted using an encryption algorithm. Symmetric or asymmetric encryption algorithms are usually used to ensure that only the receiver can decrypt; key management, secure management of the keys required for encryption and decryption. For asymmetric encryption, the sender encrypts the data using the receiver's public key, and the receiver decrypts it using the private key; secure transmission, the encrypted data is sent to the receiver through a secure transmission protocol (such as TLS / SSL) to ensure that the data is not intercepted or tampered with during transmission; data decryption, after receiving the data, the receiver uses the corresponding decryption key to decrypt the data and restore it to a plaintext state; integrity verification, during the data transmission process, hash functions or digital signatures can be used to ensure that the data has not been tampered with and verify the integrity and source of the data. Through the above steps, end-to-end encryption can effectively protect the security of sensitive data during transmission and prevent data leakage and unauthorized access.

[0068] S3, after receiving the data, the data analysis and processing unit first decrypts it, then cleans and pre-processes the data, and then retrieves the data from the cloud network database. Finally, it uses machine learning algorithms and statistical methods to combine the processed data with the data retrieved from the database for real-time analysis;

[0069] Specifically, data cleaning and preprocessing include data deduplication, data filling and data standardization.

[0070] Specifically, during the real-time data analysis process, we first need to extract frequency domain features through Fourier transform, or extract trend and periodic features through time series analysis. Then, based on whether the data is labeled, we select the corresponding algorithm for learning, and then use the trial model for training. Finally, we use the trained model to identify data points that are significantly different from the normal pattern.

[0071] It should be noted that data cleaning is to deal with missing data, outliers and noise to ensure the quality of data and extract useful features from the original data, such as extracting frequency domain features through Fourier transform, or extracting trend and periodic features through time series analysis.

[0072] Furthermore, the choice of algorithms includes classification or regression algorithms and clustering algorithms;

[0073] It should be noted that classification or regression algorithms include decision trees, random forests, support vector machines, and are used for supervised learning, where the model is trained using labeled data to predict class labels (classification) or continuous values ​​(regression), where:

[0074] Decision tree, a tree-structured model, recursively divides the data set into different subsets and selects the optimal split point based on features. Decision trees are easy to understand and explain, but may overfit;

[0075] Random forest, an ensemble learning method, improves the accuracy and stability of predictions by building multiple decision trees and combining their outputs. Random forests have a certain resistance to overfitting;

[0076] Support vector machine, a powerful algorithm for classification and regression, maximizes the interval between different categories by finding the best hyperplane. It is suitable for high-dimensional data and has significant effects, but it is sensitive to parameter selection and kernel function.

[0077] Clustering algorithms include K-means clustering algorithm, DBSCAN algorithm and convolutional neural network algorithm.

[0078] It should be noted that clustering algorithms are used for unsupervised learning. The goal is to group data so that the data points in the same group are similar, while the data points between different groups are quite different.

[0079] K-means clustering algorithm, a simple and commonly used clustering algorithm, divides the data into K clusters and iteratively adjusts the center points of the clusters until convergence. It is suitable for spherical clusters and has high computational efficiency, but is sensitive to noise and outliers;

[0080] DBSCAN algorithm, a density-based clustering algorithm, can find clusters of any shape and effectively handle noisy data. It identifies core points, boundary points, and noise points by defining the density of points, and is suitable for processing large-scale data sets.

[0081] Convolutional neural networks can be used for feature extraction and clustering. The spatial features of data are extracted through the convolution layer, and then the data is grouped in combination with the clustering algorithm.

[0082] S4, storing the processed data in a cloud network database and performing data backup at the same time;

[0083] It should be noted that data backup can avoid data loss. At the same time, multiple methods can be used to obtain original data to ensure that the data is more complete.

[0084] S5, visualizes real-time data and analysis results through dashboards and charts, and sets threshold areas for comparison.

[0085] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A traffic infrastructure safety monitoring system, characterized in that: It includes a data acquisition unit, a data transmission unit, a data analysis and processing unit, a cloud network database, a storage unit and a visualization unit; The data acquisition unit forms a monitoring network through a variety of sensors and cameras, and is used to collect traffic information through the sensors and cameras; The data transmission unit establishes a wireless communication network at each sensor and camera for real-time data transmission; The data analysis and processing unit receives the monitoring data transmitted by the data transmission unit, cleans and preprocesses the data in combination with the cloud network data, and then performs modeling analysis on the cleaned and preprocessed data through modeling; The cloud network database is used to provide traffic related information published on social media; The storage unit is used for storing data; The visualization unit displays the analysis results in a graphical form to assist in determining the safety status of traffic facilities.

2. A traffic infrastructure safety monitoring system according to claim 1, characterized in that: The types of sensors include environmental sensors, structural health monitoring sensors, and vehicle monitors; The environmental sensors include temperature and humidity sensors, rainfall sensors, wind speed sensors and infrared sensors, which are used to monitor meteorological conditions, including temperature, humidity and wind speed; The structural health monitoring sensors include strain sensors, displacement sensors and piezoelectric sensors, which are used to monitor the structural status of transportation infrastructure, including strain, displacement and vibration; The vehicle monitor uses a millimeter wave radar, which transmits electromagnetic waves and receives reflected signals to count vehicle flow, vehicle speed and vehicle type classification.

3. A traffic infrastructure safety monitoring system according to claim 1, characterized in that: The data analysis and processing unit includes a data deduplication module, a data filling module and a data standardization module, and performs data cleaning and preprocessing through the deduplication module, the data filling module and the data standardization module.

4. A traffic infrastructure safety monitoring system according to claim 3, characterized in that: The deduplication module is used to ensure the uniqueness of each piece of data by removing redundant information; The data filling module fills the missing data using interpolation and mean filling methods. The interpolation method estimates the missing data by using the relationship between existing data points, and the mean filling method replaces the missing value with the mean of other observations in the data set. The data standardization module is used to unify the formats of data from different sources.

5. A method for monitoring the safety of traffic infrastructure, using a monitoring system according to any one of claims 1 to 4, characterized in that: The monitoring method comprises the following steps: S1, first establish a data acquisition unit and a data transmission unit, set up sensors and cameras for information acquisition in the area to be monitored, and after setting up the sensors and cameras, establish a wireless network communication for the transmission of the collected data; S2, wireless network communication transmits the collected data to the data analysis and processing unit in real time, and encrypts the data during the transmission process; S3, after receiving the data, the data analysis and processing unit first decrypts it, then cleans and pre-processes the data, and then retrieves the data from the cloud network database. Finally, it uses machine learning algorithms and statistical methods to combine the processed data with the data retrieved from the database for real-time analysis; S4, storing the processed data in a cloud network database and performing data backup at the same time; S5, visualizes real-time data and analysis results through dashboards and charts, and sets threshold areas for comparison.

6. A method for monitoring traffic infrastructure safety according to claim 5, characterized in that: The setting type of the sensor is selected according to the data to be collected, and the time interval error of each sensor transmitting data is 0.5-0.8s, and the transmission time is calibrated every 5-7 days.

7. A method for monitoring traffic infrastructure safety according to claim 5, characterized in that: The data is transmitted using end-to-end encryption, and the data is always kept encrypted during the entire process of being sent from the sensor to the receiving end.

8. A method for monitoring traffic infrastructure safety according to claim 5, characterized in that: The data cleaning and preprocessing include data deduplication, data filling and data standardization.

9. A method for monitoring traffic infrastructure safety according to claim 5, characterized in that: During the real-time analysis of the data, the frequency domain features must first be extracted through Fourier transform, or trend and periodic features must be extracted through time series analysis. Then, based on whether the data is labeled, the corresponding algorithm is selected for learning, and then the real-time model is used for training. Finally, the trained model is used to identify data points that are significantly different from the normal pattern.

10. A method for monitoring traffic infrastructure safety according to claim 9, characterized in that: The selection of the algorithm includes classification or regression algorithm and clustering algorithm; The classification or regression algorithm includes decision tree, random forest, support vector machine; The clustering algorithms include K-means clustering algorithm, DBSCAN algorithm and convolutional neural network algorithm.

Images