Bridge cluster safety monitoring and early warning method and system based on Beidou high-precision positioning

Abstract

The invention belongs to the technical field of bridge monitoring, and discloses a bridge cluster safety monitoring and early warning method and system based on Beidou high-precision positioning. The method comprises the following steps: constructing an artificial intelligence model based on a cloud data center; acquiring real-time positioning data and real-time sensor data based on a bridge data acquisition device; matrix conversion is carried out based on the cloud data center; performing bridge cluster anomaly analysis by using the bridge cluster anomaly analysis model; generating a real-time bridge cluster safety early warning signal; acquiring a real-time abnormal bridge satellite remote sensing image; performing bridge safety detection by using the bridge safety detection model; generating a real-time abnormal bridge safety early warning signal; performing bridge safety optimization by using the bridge safety optimization model; and sending the real-time bridge safety optimization strategy to a bridge data acquisition device. According to the invention, the problems of low efficiency, high cost, poor functionality, low positioning precision and lack of intelligence in the prior art are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of bridge monitoring, and in particular relates to a bridge cluster safety monitoring and early warning method and system based on Beidou high-precision positioning. Background Art

[0002] As an important part of transportation infrastructure, the safety of bridges plays a vital role in protecting people's lives and property and promoting economic development. With the increasing number of bridge construction and the extension of bridge service life, bridge safety monitoring and early warning have become an increasingly important issue. The existing bridge safety monitoring methods mainly have the following problems:

[0003] 1) Inefficiency and high cost: Existing bridge safety monitoring methods mainly rely on manual inspections and regular testing, which is not only inefficient but also costly;

[0004] 2) Poor functionality: Existing bridge safety monitoring methods cannot achieve real-time monitoring and early warning of bridge clusters. They lack real-time monitoring and early warning capabilities and are difficult to detect and deal with potential safety hazards in a timely manner.

[0005] 3) Low positioning accuracy: The existing bridge safety monitoring methods have low accuracy in positioning data collection and analysis, making it difficult to accurately determine the exact location of the bridge;

[0006] 4) Lack of intelligence: Existing bridge safety monitoring methods lack intelligence and cannot automatically provide bridge safety optimization strategies based on real-time data, resulting in low practicality. Summary of the invention

[0007] In order to solve the problems of low efficiency, high cost, poor functionality, low positioning accuracy and lack of intelligence in the prior art, the present invention aims to provide a bridge cluster safety monitoring and early warning method and system based on Beidou high-precision positioning.

[0008] The technical solution adopted by the present invention is:

[0009] A bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning includes the following steps:

[0010] Based on the cloud data center, using artificial intelligence algorithms, we build bridge cluster anomaly analysis models, bridge safety detection models, and bridge safety optimization models;

[0011] Based on the bridge data collection device, the real-time positioning data and real-time sensor data of the corresponding bridge are collected and uploaded to the cloud data center;

[0012] Based on the cloud data center, the real-time sensor data of several bridges are matrix-converted according to the corresponding real-time positioning data to obtain the real-time sensor data matrix of the bridge cluster;

[0013] Based on the cloud data center, according to the real-time sensor data matrix, the bridge cluster anomaly analysis model is used to perform bridge cluster anomaly analysis and obtain real-time bridge cluster anomaly analysis results;

[0014] Based on the cloud data center, if there are real-time bridge abnormal targets in the real-time bridge cluster abnormal analysis results, a real-time bridge cluster safety warning signal is generated according to the corresponding real-time positioning data;

[0015] Based on the cloud data center, according to the real-time positioning data of the real-time abnormal bridge targets, the corresponding real-time abnormal bridge satellite remote sensing images are collected;

[0016] Based on the cloud data center, according to the real-time abnormal bridge satellite remote sensing images, the bridge safety detection model is used to perform bridge safety detection and obtain real-time bridge safety detection results;

[0017] Based on the cloud data center, if the real-time bridge safety detection result shows that there is a bridge safety risk, a real-time abnormal bridge safety warning signal is generated according to the corresponding abnormal real-time positioning data;

[0018] Based on the cloud data center, according to the real-time bridge safety detection results with bridge safety risks, the bridge safety optimization model is used to optimize the bridge safety and obtain the real-time bridge safety optimization strategy;

[0019] Based on the cloud data center, the real-time bridge safety optimization strategy is sent to the bridge data acquisition device corresponding to the bridge safety risk in the real-time bridge safety detection result.

[0020] Furthermore, based on the cloud data center, using artificial intelligence algorithms, a bridge cluster anomaly analysis model, a bridge safety detection model, and a bridge safety optimization model are constructed, including the following steps:

[0021] Based on the cloud data center, historical positioning data and historical sensor data of several bridges are collected and preprocessed to obtain several preprocessed historical sensor data and preprocessed historical positioning data;

[0022] According to the corresponding pre-processed historical positioning data, a number of pre-processed historical sensor data are matrix-converted to obtain the historical sensor data matrix of the bridge cluster;

[0023] Perform data dimensionality reduction on the historical sensor data matrix to obtain a principal component set and a historical sensor data matrix after dimensionality reduction;

[0024] Based on the historical sensor data matrix after dimensionality reduction, a deep learning algorithm is used to build a bridge cluster anomaly analysis model and generate several historical bridge cluster anomaly analysis results;

[0025] According to the corresponding pre-processed historical positioning data, the historical abnormal bridge satellite remote sensing images of several bridges are collected, and the samples are expanded and pre-processed to obtain the expanded image sample set;

[0026] Based on the expanded image sample set, an image recognition algorithm is used to build a bridge safety detection model and generate several historical bridge safety detection results;

[0027] Based on several historical bridge safety inspection results, a bridge safety optimization model is constructed using a reinforcement learning algorithm, and several generated historical bridge safety optimization experiences are stored in an experience replay pool.

[0028] Furthermore, the PCA method is used to perform data dimensionality reduction on the historical sensor data matrix to obtain a principal component set and a historical sensor data matrix after dimensionality reduction.

[0029] Furthermore, the bridge cluster anomaly analysis model is constructed based on the CNN-BiLSTM algorithm, and the bridge cluster anomaly analysis model includes a matrix feature extraction module constructed based on the CNN algorithm and a bridge cluster anomaly analysis module constructed based on the BiLSTM algorithm;

[0030] The bridge safety detection model is built based on the GSCNN-ShuffleNet algorithm, and the bridge cluster anomaly analysis model includes an image segmentation module built based on the GSCNN algorithm and a bridge safety detection module built based on the ShuffleNet algorithm;

[0031] The bridge safety optimization model is built based on the DQN algorithm, and the bridge safety optimization model includes a deep Q network, an intelligent agent, and an experience replay pool.

[0032] Furthermore, based on the cloud data center, according to the corresponding real-time positioning data, the real-time sensor data of several bridges are matrix-converted to obtain the real-time sensor data matrix of the bridge cluster, including the following steps:

[0033] Based on the cloud data center, the real-time positioning data of several bridges are clustered with the bridge location as the clustering target using the FCM algorithm to obtain several real-time positioning data clusters.

[0034] Extracting the real-time sensor data corresponding to each real-time positioning data in the same real-time positioning data clustering cluster, and obtaining the real-time sensor data clustering clusters of the same monitoring area;

[0035] The real-time sensor data of the same monitoring area are clustered and transformed into a matrix to obtain the real-time sensor data matrix of the bridge cluster in the same monitoring area.

[0036] Furthermore, based on the cloud data center and according to the real-time sensor data matrix, the bridge cluster anomaly analysis model is used to perform bridge cluster anomaly analysis to obtain real-time bridge cluster anomaly analysis results, including the following steps:

[0037] Based on the cloud data center, the real-time sensor data matrix is ​​subjected to data dimension reduction according to the principal component set to obtain the real-time sensor data matrix after dimension reduction;

[0038] According to the real-time sensor data matrix, the matrix feature extraction module of the bridge cluster anomaly analysis model is used to extract matrix features and obtain real-time matrix features;

[0039] According to the real-time matrix characteristics, the bridge cluster anomaly analysis module of the bridge cluster anomaly analysis model is used to perform bridge cluster anomaly analysis to obtain real-time bridge cluster anomaly analysis results.

[0040] Furthermore, based on the cloud data center, according to the real-time abnormal bridge satellite remote sensing image, using the bridge safety detection model, bridge safety detection is performed to obtain real-time bridge safety detection results, including the following steps:

[0041] Based on the cloud data center, the real-time abnormal bridge satellite remote sensing image is preprocessed to obtain the preprocessed real-time abnormal bridge satellite remote sensing image;

[0042] According to the real-time abnormal bridge satellite remote sensing image after preprocessing, the image segmentation module of the bridge safety detection model is used to perform image segmentation to obtain a real-time bridge segmentation image;

[0043] According to the real-time bridge segmentation image, the bridge safety detection module of the bridge safety detection model is used to perform bridge safety detection to obtain real-time bridge safety detection results.

[0044] Furthermore, based on the cloud data center, according to the real-time bridge safety detection results with bridge safety risks, the bridge safety optimization model is used to optimize the bridge safety and obtain the real-time bridge safety optimization strategy, which includes the following steps:

[0045] Based on the cloud data center, according to the real-time bridge safety detection results with bridge safety risks, the corresponding historical bridge safety optimization experience is retrieved in the experience playback pool to obtain several matching historical bridge safety optimization experiences;

[0046] According to a number of matching historical bridge safety optimization experiences, the action space of the bridge safety optimization model is updated to obtain an updated action space;

[0047] According to the real-time bridge safety detection results of the bridge with safety risks, the state space of the bridge safety optimization model is updated to obtain an updated state space;

[0048] Use the updated state space as the input of the deep Q network and use the agent to generate the Q value of each possible action in the updated action space;

[0049] Use the reward function to obtain the reward value of each possible action in the updated action space, and update the Q value of the possible action according to the reward value to obtain the updated Q value of each possible action;

[0050] Repeat the above Q-value updating steps until the predetermined number of iterations is reached, and use a greedy strategy to select the possible action with the highest updated Q-value in each iteration as the execution action;

[0051] Several execution actions are integrated to obtain a real-time bridge safety optimization strategy.

[0052] Furthermore, it also includes generating keys and registering identities for bridge data collection devices based on a trusted institution, obtaining corresponding public-private key pairs and signature information, returning the private key and signature information in the public-private key pairs to the corresponding bridge data collection devices, and publishing the public key in the public-private key pairs to a cloud data center;

[0053] The private key is used to encrypt the real-time positioning data and real-time sensor data collected by the bridge data collection device to obtain encrypted real-time data;

[0054] The signature information is used to sign the encrypted real-time data to obtain real-time signature data.

[0055] A bridge cluster safety monitoring and early warning system based on Beidou high-precision positioning is used to implement a bridge cluster safety monitoring and early warning method. The system includes a cloud data center, a trusted organization, and several bridge data acquisition devices;

[0056] The cloud data center is respectively connected to the trusted institution and a plurality of bridge data acquisition devices in communication, and the cloud data center includes a model building unit, a signature authentication and decryption unit, a matrix conversion unit, a bridge cluster abnormality analysis unit, a first safety warning unit, an image acquisition unit, a bridge safety detection unit, a second safety warning unit, a bridge safety optimization and a data sending unit which are connected in sequence;

[0057] The trusted institutions are respectively connected in communication with a plurality of bridge data collection devices;

[0058] The bridge data acquisition device includes a main control module, a data encryption and signature module, a Beidou high-precision positioning module, a sensor module and a data upload module. The main control module is electrically connected to the data encryption and signature module, the Beidou high-precision positioning module, the sensor module and the data upload module, and the data upload module is communicated with the cloud data center and the trusted organization respectively.

[0059] The beneficial effects of the present invention are:

[0060] The present invention provides a bridge cluster safety monitoring and early warning method and system based on Beidou high-precision positioning, which integrates Beidou high-precision positioning technology and utilizes the high-precision positioning capability of the Beidou satellite navigation system to achieve real-time positioning and monitoring of bridge clusters, improve the accuracy of positioning data collection, and provide accurate basic data for subsequent safety analysis and early warning; construct a bridge cluster abnormality analysis model, a bridge safety detection model, and a bridge safety optimization model to achieve intelligent analysis and prediction of bridge safety conditions, improve functionality, automatically identify potential safety risks, and provide customized bridge safety optimization strategies, thereby improving intelligence and practicality; through real-time collection of bridge data, real-time monitoring and early warning of bridge clusters are achieved, potential safety hazards are discovered and handled in a timely manner, and real-time and early warning capabilities are improved; centralized data management and analysis are achieved through a cloud data center, scalability and flexibility are improved, support for large-scale bridge cluster monitoring and analysis, reduce manual intervention, and improve efficiency.

[0061] Other beneficial effects of the present invention will be further described in the specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a flowchart of the bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning in the present invention.

[0063] Figure 2 It is a structural block diagram of the bridge cluster safety monitoring and early warning system based on Beidou high-precision positioning in the present invention. DETAILED DESCRIPTION

[0064] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.

[0065] Embodiment 1:

[0066] like Figure 1 As shown, this embodiment provides a bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning, including the following steps:

[0067] S1: Based on the cloud data center, using artificial intelligence algorithms, build a bridge cluster anomaly analysis model, a bridge safety detection model, and a bridge safety optimization model, including the following steps:

[0068] S1-1: Based on the cloud data center, historical positioning data and historical sensor data of several bridges are collected and preprocessed to obtain several preprocessed historical sensor data and preprocessed historical positioning data;

[0069] S1-2: According to the corresponding pre-processed historical positioning data, a number of pre-processed historical sensor data are matrix-converted to obtain a historical sensor data matrix of the bridge cluster;

[0070] S1-3: Use the principal component analysis (PCA) method to reduce the dimension of the historical sensor data matrix to obtain a principal component set and a historical sensor data matrix after dimension reduction;

[0071] S1-4: Based on the historical sensor data matrix after dimensionality reduction, a deep learning algorithm is used to build a bridge cluster anomaly analysis model and generate several historical bridge cluster anomaly analysis results;

[0072] The bridge cluster anomaly analysis model is constructed based on the Convolutional Neural Networks (CNN)-Bi-directional Long Short-Term Memory (BiLSTM) algorithm, and the bridge cluster anomaly analysis model includes a matrix feature extraction module constructed based on the CNN algorithm and a bridge cluster anomaly analysis module constructed based on the BiLSTM algorithm;

[0073] S1-5: According to the corresponding pre-processed historical positioning data, the historical abnormal bridge satellite remote sensing images of several bridges are collected, and the samples are expanded and pre-processed to obtain the expanded image sample set;

[0074] S1-6: Based on the expanded image sample set, use the image recognition algorithm to build a bridge safety detection model and generate several historical bridge safety detection results;

[0075] The bridge safety detection model is built based on the Gated-Shape Convolutional Neural Networks (GSCNN)-ShuffleNet algorithm, and the bridge cluster anomaly analysis model includes an image segmentation module built based on the GSCNN algorithm and a bridge safety detection module built based on the ShuffleNet algorithm;

[0076] The image segmentation module includes a rule flow module, a shape flow module and a fusion module. The rule flow module adopts a classic convolutional neural network structure for processing semantic region information. The shape flow module consists of multiple gated convolutional layers. The low-level feature map in the rule flow module is used to process the region boundary information. The outputs of the two modules are sent to a fusion module. In the fusion module, the outputs of the rule flow module and the shape flow module are combined using the Atrous Special Pyramid Pooling (ASPP) structure. ASPP helps preserve multi-scale context information. Finally, the fusion module generates a semantic region of an object with a defined boundary to obtain an accurately segmented semantic segmentation image.

[0077] The bridge safety detection module uses multiple group convolution layers to reduce the amount of calculation and parameters by several times when the segmented images have the same input and output dimensions, thereby achieving the purpose of lightweight network. The first group convolution layer extracts the segmented image features. The feature fusion layer obtains the fusion features by multiplying the data weight array with the input segmented image feature data. The Shuffle layer randomly shuffles the feature map of the fusion features, so that the information between the groups of the second group convolution layer can flow between each other without any additional calculation cost, thereby improving the recognition efficiency and accuracy of bridge safety detection. The output layer outputs the corresponding bridge safety detection results.

[0078] S1-7: Based on several historical bridge safety inspection results, a reinforcement learning algorithm is used to construct a bridge safety optimization model, and several generated historical bridge safety optimization experiences are stored in the experience playback pool, including the following steps:

[0079] S1-7-1: Generate the bridge safety optimization problem as the simulation environment of the DQN algorithm, and build the intelligent agent and experience replay pool;

[0080] S1-7-2: Based on each bridge state type corresponding to the historical bridge safety inspection results, define the state space of the DQN algorithm, and the parameters of the state space correspond to each bridge state;

[0081] S1-7-3: Define the action space of the DQN algorithm according to the actions that need to be output by the bridge safety optimization strategy;

[0082] S1-7-4: Based on the possible impact of each action in the action space, define the reward function of the DQN algorithm to evaluate the quality or impact of the action;

[0083] S1-7-5: Construct the input layer, several hidden layers, and output layer of the deep Q network, connect the input layer to the state space, and connect the output layer to the action space;

[0084] S1-7-6: Based on the state space, action space and reward function, according to several historical abnormal traffic detection results and corresponding several historical key dimension features, the deep Q network and intelligent agent are optimized and trained to build a bridge safety optimization model, and several historical bridge safety optimization experiences are generated;

[0085] S1-7-7: Store several historical bridge safety optimization experiences into the experience playback pool of the bridge safety optimization model;

[0086] The bridge safety optimization model is built based on the Deep Q Network (DQN) algorithm, and the bridge safety optimization model includes a deep Q network, an intelligent agent, and an experience replay pool;

[0087] S1.1: Based on the trusted institution, key generation and identity registration are performed on the bridge data acquisition device to obtain the corresponding public-private key pair and signature information, the private key and signature information in the public-private key pair are returned to the corresponding bridge data acquisition device, and the public key in the public-private key pair is published to the cloud data center, including the following steps:

[0088] S1.1-1: Initialize the key based on the trusted institution to obtain the public parameters, master key and initial key;

[0089] The formula is:

[0090] GP={g,g1,g a ,e(g,g) a ,H1,H2,H3,H4,H5,H6}

[0091] PK={g,g1,g a ,e(g,g) a ,H u}

[0092] MSK = {g a ,a}

[0093] Where GP is the public parameter; MSK is the master key; PK is the initial key; a is the integer domain Z p Random numbers; H1, H2, H3, H4, H5, H6, H u All are target hash functions; g, g1, g a are all random numbers generated by generators in the cyclic group G; e(g,g) a is a bilinear map of random numbers g;

[0094] S1.1-2: According to the attribute information V in the device information u, public parameters, master key and initial key, use asymmetric encryption algorithm to generate keys for bridge data acquisition devices and obtain corresponding public-private key pairs;

[0095] The formula is:

[0096] SK u = {MSK,V u ,K=g a g ab ,L u =g b ,(K' u =H3(V u ) b )}

[0097]

[0098] In the formula, SK u is the private key of the bridge data acquisition device u; b is the integer domain Z p Random number; L u , K' u is the private key parameter of the bridge data acquisition device u; H3 is the target hash function of the public parameter GP; u is the indicator of the bridge data acquisition device; MSK is the master key; PK is the initial key; PK u is the public key of the bridge data acquisition device u; g b , g a , g ab V is a random number generated by a generator in the cyclic group G; u is the attribute information of the bridge data collection device u;

[0099] S1.1-3: According to the device ID in the device information, use the digital identity authentication method to register the bridge data collection device and obtain the corresponding signature information;

[0100] The formula is:

[0101]

[0102] Where k′ is a random number; K u is the registration parameter of the bridge data acquisition device u; KID u KID is the registration ID of the bridge data collection device u; u and the corresponding K u The signature information of the bridge data acquisition device {K u ,KID u}; H1 is the target hash function; ID u is the device ID of the bridge data collection device u; is the prime order; P is the base point of the prime domain; mod(*) is the remainder function;

[0103] S1.1-4: Return the private key and signature information in the public-private key pair to the corresponding bridge data acquisition device, and publish the public key in the public-private key pair to the cloud data center;

[0104] S2: Based on the bridge data collection device, the real-time positioning data and real-time sensor data of the corresponding bridge are collected and uploaded to the cloud data center, including the following steps:

[0105] S2-1: Based on the bridge data collection device, collect the real-time positioning data and real-time sensor data of the corresponding bridge;

[0106] S2-2: Encrypt the real-time positioning data and real-time sensor data of the bridge according to the private key to obtain encrypted real-time data;

[0107] The formula is:

[0108] M u1 =E(SK u ,m u )

[0109] Where M u1 is the encrypted real-time data; E(*) is the asymmetric encryption function; m u Real-time positioning data and real-time sensor data for bridges; SK u is the private key of the data collection device u;

[0110] S2-3: Sign the encrypted real-time data according to the signature information to obtain real-time signature data;

[0111] The formula is:

[0112]

[0113] In the formula, r' is a random number; is the prime order; P is the base point of the prime domain; H2 is the target hash function; K u For the signature information {K u ,KID u}Registration parameters of bridge data acquisition device u; KID u For the signature information {K u ,KID u}Registration ID of bridge data acquisition device u; ID u is the device ID of the bridge data collection device u; the real-time signature data is {ID u ,γ'={K u ,R u ,M u1 ,Bu}}; R u ,B u ,γ' are the signature parameters of the bridge data acquisition device u;

[0114] S2-4: Upload the encrypted real-time data and the real-time signature data to the cloud data center;

[0115] Combining asymmetric encryption technology and digital identity authentication technology to ensure the security and integrity of data during transmission, while improving the protection level of cloud data centers;

[0116] S3: Based on the cloud data center, according to the corresponding real-time positioning data, the real-time sensor data of several bridges are matrix-converted to obtain the real-time sensor data matrix of the bridge cluster, including the following steps:

[0117] S3-0: Based on the cloud data center, the trusted organization is called to perform signature verification on the real-time signature data. After the signature verification is passed, the encrypted real-time data is decrypted according to the public key to obtain the decrypted real-time data;

[0118] The formula is:

[0119] β u B u P=β u H2(R u ,ID u ,M u1 ,K u )R u +β u K u +β u H1(ID u ,K u )PK u

[0120] In the formula, β u PK is the signature verification parameter of the bridge data acquisition device u; u is the public key of the bridge data collection device u; if the left formula is equal to the right formula, the signature verification is successful;

[0121] m u2 =E - (PK u |RR2,M u1 )

[0122] In the formula, m u2 Decrypted real-time data, including real-time positioning data and real-time sensor data; E - (*) is an asymmetric decryption function; M u1 It is the encrypted real-time data; PK uis the public key of the bridge data collection device u;

[0123] S3-1: Based on the cloud data center, with the bridge location as the clustering target, the Fuzzy C-Means clustering algorithm (FCM) is used to cluster the real-time positioning data of several bridges to obtain several real-time positioning data clusters;

[0124] S3-2: extracting the real-time sensor data corresponding to each real-time positioning data in the same real-time positioning data clustering cluster, and obtaining the real-time sensor data clustering clusters of the same monitoring area;

[0125] S3-3: cluster the real-time sensor data of the same monitoring area and perform matrix conversion to obtain the real-time sensor data matrix of the bridge cluster in the same monitoring area;

[0126] S4: Based on the cloud data center, according to the real-time sensor data matrix, using the bridge cluster anomaly analysis model, bridge cluster anomaly analysis is performed to obtain real-time bridge cluster anomaly analysis results, including the following steps:

[0127] S4-1: Based on the cloud data center, the real-time sensor data matrix is ​​subjected to data dimension reduction according to the principal component set to obtain the real-time sensor data matrix after dimension reduction;

[0128] S4-2: According to the real-time sensor data matrix, use the matrix feature extraction module of the bridge cluster anomaly analysis model to extract matrix features and obtain real-time matrix features;

[0129] S4-3: According to the real-time matrix characteristics, the bridge cluster anomaly analysis module of the bridge cluster anomaly analysis model is used to perform bridge cluster anomaly analysis to obtain real-time bridge cluster anomaly analysis results;

[0130] S5: Based on the cloud data center, if there is a real-time bridge abnormality target in the real-time bridge cluster abnormality analysis result, a real-time bridge cluster safety warning signal is generated according to the corresponding real-time positioning data;

[0131] S6: Based on the cloud data center, according to the real-time positioning data of the real-time abnormal bridge target, the corresponding real-time abnormal bridge satellite remote sensing image is collected;

[0132] S7: Based on the cloud data center, according to the real-time abnormal bridge satellite remote sensing image, using the bridge safety detection model, bridge safety detection is performed to obtain real-time bridge safety detection results, including the following steps:

[0133] S7-1: Based on the cloud data center, the real-time abnormal bridge satellite remote sensing image is preprocessed to obtain the preprocessed real-time abnormal bridge satellite remote sensing image;

[0134] S7-2: Based on the pre-processed real-time abnormal bridge satellite remote sensing image, use the image segmentation module of the bridge safety detection model to perform image segmentation to obtain a real-time bridge segmentation image;

[0135] S7-3: Based on the real-time bridge segmentation image, using the bridge safety detection module of the bridge safety detection model, bridge safety detection is performed to obtain a real-time bridge safety detection result;

[0136] S8: Based on the cloud data center, if the real-time bridge safety detection result indicates that there is a bridge safety risk, a real-time abnormal bridge safety warning signal is generated according to the corresponding abnormal real-time positioning data;

[0137] S9: Based on the cloud data center, according to the real-time bridge safety detection results with bridge safety risks, the bridge safety optimization model is used to optimize the bridge safety and obtain the real-time bridge safety optimization strategy, including the following steps:

[0138] S9-1: Based on the cloud data center, according to the real-time bridge safety detection results with bridge safety risks, the corresponding historical bridge safety optimization experience is retrieved in the experience playback pool to obtain several matching historical bridge safety optimization experiences;

[0139] S9-2: Based on some matching historical bridge safety optimization experiences, the action space of the bridge safety optimization model is updated to obtain the updated action space A'=[a'1,...,a' j" ,...,a' I ], where a' j" is the updated j-th "action value", j" is the action indicator, and I is the total number of action space dimensions;

[0140] S9-3: Based on the real-time bridge safety detection results with bridge safety risks, the state space of the bridge safety optimization model is updated to obtain an updated state space S' = [s'1, ..., s' i" ,...,s' I' ], where s' i" is the updated i'th state value, i' is the state indicator, and I' is the total number of state space dimensions;

[0141] S9-4: Use the updated state space as the input of the deep Q network and use the agent to generate the Q value of each possible action in the updated action space;

[0142] S9-5: Use the reward function to obtain the reward value of each possible action in the updated action space, and update the Q value of the possible action according to the reward value to obtain the updated Q value of each possible action;

[0143] The formula is:

[0144] Q(s' p' ,a' p' )=(1-α")·Q(s p' ,a p' )+α"·(R(s p' ,a p' ,s' p' )+γ·Q max (s p' ,a p' ))

[0145] In the formula, Q(s' p' ,a' p' ) is the updated state value s' p' and the updated action value a' p' The corresponding updated Q value; Q(s p' ,a p' ) is the state value s p' and action value a p' The corresponding predicted Q value; α" is the learning rate; Q max (s p' ,a p' ) is the highest predicted Q value; p' is the comprehensive indicator; γ is the update parameter;

[0146] S9-6: Repeat the above Q value update steps until the predetermined number of iterations is reached, and use the greedy strategy to select the possible action with the highest updated Q value in each iteration as the execution action;

[0147] S9-7: Integrate several execution actions to obtain a real-time bridge safety optimization strategy;

[0148] S10: Based on the cloud data center, the real-time bridge safety optimization strategy is sent to the bridge data acquisition device corresponding to the real-time bridge safety detection result that there is a bridge safety risk.

[0149] Embodiment 2:

[0150] like Figure 2 As shown, this embodiment provides a bridge cluster safety monitoring and early warning system based on Beidou high-precision positioning, which is used to implement a bridge cluster safety monitoring and early warning method. The system includes a cloud data center, a trusted organization, and several bridge data collection devices;

[0151] The cloud data center is respectively connected to the trusted institution and a plurality of bridge data acquisition devices in communication, and the cloud data center includes a model building unit, a signature authentication and decryption unit, a matrix conversion unit, a bridge cluster abnormality analysis unit, a first safety warning unit, an image acquisition unit, a bridge safety detection unit, a second safety warning unit, a bridge safety optimization and a data sending unit which are connected in sequence;

[0152] A model building unit, used to build a bridge cluster anomaly analysis model, a bridge safety detection model, and a bridge safety optimization model using an artificial intelligence algorithm;

[0153] The signature authentication and decryption unit is used to call a trusted institution based on the cloud data center to perform signature verification on the real-time signature data. After the signature verification is passed, the encrypted real-time data is decrypted according to the public key to obtain the decrypted real-time data;

[0154] A matrix conversion unit, used to perform matrix conversion on the real-time sensor data of a number of bridges according to the corresponding real-time positioning data, to obtain a real-time sensor data matrix of the bridge cluster;

[0155] A bridge cluster anomaly analysis unit is used to perform bridge cluster anomaly analysis based on a real-time sensor data matrix and a bridge cluster anomaly analysis model to obtain a real-time bridge cluster anomaly analysis result;

[0156] The first safety warning unit is used to generate a real-time bridge cluster safety warning signal according to the corresponding real-time positioning data when there is a real-time bridge abnormality target in the real-time bridge cluster abnormality analysis result;

[0157] An image acquisition unit is used to acquire corresponding real-time abnormal bridge satellite remote sensing images according to real-time positioning data of real-time abnormal bridge targets;

[0158] A bridge safety detection unit is used to perform bridge safety detection based on real-time abnormal bridge satellite remote sensing images and use a bridge safety detection model to obtain real-time bridge safety detection results;

[0159] The second safety warning unit is used to generate a real-time abnormal bridge safety warning signal according to the corresponding abnormal real-time positioning data when the real-time bridge safety detection result indicates that there is a bridge safety risk;

[0160] Bridge safety optimization is used to optimize bridge safety based on the real-time bridge safety detection results with bridge safety risks, using the bridge safety optimization model to obtain a real-time bridge safety optimization strategy;

[0161] A data sending unit, used to send the real-time bridge safety optimization strategy to a bridge data acquisition device corresponding to the presence of a bridge safety risk in a real-time bridge safety detection result;

[0162] The trusted institutions are respectively connected in communication with a plurality of bridge data collection devices;

[0163] The trusted institution is used to generate keys and register identities for the bridge data collection device, obtain the corresponding public-private key pair and signature information, return the private key and signature information in the public-private key pair to the corresponding bridge data collection device, and publish the public key in the public-private key pair to the cloud data center;

[0164] The bridge data acquisition device includes a main control module, a data encryption and signature module, a Beidou high-precision positioning module, a sensor module and a data upload module. The main control module is electrically connected to the data encryption and signature module, the Beidou high-precision positioning module, the sensor module and the data upload module respectively, and the data upload module is communicated with the cloud data center and the trusted institution respectively;

[0165] The main control module is used to control the normal operation of other modules;

[0166] Beidou high-precision positioning module, used to collect real-time positioning data of the corresponding bridge;

[0167] The sensor module collects real-time sensor data of the corresponding bridge, including real-time pressure data, real-time deformation data, real-time environmental data and other monitoring data of the bridge;

[0168] The data encryption and signature module is used to encrypt the real-time positioning data and real-time sensor data of the bridge according to the private key to obtain the encrypted real-time data; and to sign the encrypted real-time data according to the signature information to obtain the real-time signature data;

[0169] The data upload module is used to upload the encrypted real-time data and real-time signature data to the cloud data center.

[0170] The present invention provides a bridge cluster safety monitoring and early warning method and system based on Beidou high-precision positioning, which integrates Beidou high-precision positioning technology and utilizes the high-precision positioning capability of the Beidou satellite navigation system to achieve real-time positioning and monitoring of bridge clusters, improve the accuracy of positioning data collection, and provide accurate basic data for subsequent safety analysis and early warning; construct a bridge cluster abnormality analysis model, a bridge safety detection model, and a bridge safety optimization model to achieve intelligent analysis and prediction of bridge safety conditions, improve functionality, automatically identify potential safety risks, and provide customized bridge safety optimization strategies, thereby improving intelligence and practicality; through real-time collection of bridge data, real-time monitoring and early warning of bridge clusters are achieved, potential safety hazards are discovered and handled in a timely manner, and real-time and early warning capabilities are improved; centralized data management and analysis are achieved through a cloud data center, scalability and flexibility are improved, support for large-scale bridge cluster monitoring and analysis, reduce manual intervention, and improve efficiency.

[0171] The present invention is not limited to the above optional implementations, and anyone can derive other various forms of products under the enlightenment of the present invention. The above specific implementations should not be understood as limiting the scope of protection of the present invention. The scope of protection of the present invention should be based on the definition in the claims, and the description can be used to interpret the claims.

Claims

1. A bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning, Features: The steps include: Based on the cloud data center, using artificial intelligence algorithms, we build bridge cluster anomaly analysis models, bridge safety detection models, and bridge safety optimization models; Based on the bridge data collection device, the real-time positioning data and real-time sensor data of the corresponding bridge are collected and uploaded to the cloud data center; Based on the cloud data center, the real-time sensor data of several bridges are matrix-converted according to the corresponding real-time positioning data to obtain the real-time sensor data matrix of the bridge cluster; Based on the cloud data center, according to the real-time sensor data matrix, the bridge cluster anomaly analysis model is used to perform bridge cluster anomaly analysis and obtain real-time bridge cluster anomaly analysis results; Based on the cloud data center, if there are real-time bridge abnormal targets in the real-time bridge cluster abnormal analysis results, a real-time bridge cluster safety warning signal is generated according to the corresponding real-time positioning data; Based on the cloud data center, according to the real-time positioning data of the real-time abnormal bridge targets, the corresponding real-time abnormal bridge satellite remote sensing images are collected; Based on the cloud data center, according to the real-time abnormal bridge satellite remote sensing images, the bridge safety detection model is used to perform bridge safety detection and obtain real-time bridge safety detection results; Based on the cloud data center, if the real-time bridge safety detection result shows that there is a bridge safety risk, a real-time abnormal bridge safety warning signal is generated according to the corresponding abnormal real-time positioning data; Based on the cloud data center, according to the real-time bridge safety detection results with bridge safety risks, the bridge safety optimization model is used to optimize the bridge safety and obtain the real-time bridge safety optimization strategy; Based on the cloud data center, the real-time bridge safety optimization strategy is sent to the bridge data acquisition device corresponding to the bridge safety risk in the real-time bridge safety detection result.

2. According to claim 1, a bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning, Features: Based on the cloud data center, using artificial intelligence algorithms, we build a bridge cluster anomaly analysis model, a bridge safety detection model, and a bridge safety optimization model, including the following steps: Based on the cloud data center, historical positioning data and historical sensor data of several bridges are collected and preprocessed to obtain several preprocessed historical sensor data and preprocessed historical positioning data; According to the corresponding pre-processed historical positioning data, a number of pre-processed historical sensor data are matrix-converted to obtain the historical sensor data matrix of the bridge cluster; Perform data dimensionality reduction on the historical sensor data matrix to obtain a principal component set and a historical sensor data matrix after dimensionality reduction; Based on the historical sensor data matrix after dimensionality reduction, a deep learning algorithm is used to build a bridge cluster anomaly analysis model and generate several historical bridge cluster anomaly analysis results; According to the corresponding pre-processed historical positioning data, the historical abnormal bridge satellite remote sensing images of several bridges are collected, and the samples are expanded and pre-processed to obtain the expanded image sample set; Based on the expanded image sample set, an image recognition algorithm is used to build a bridge safety detection model and generate several historical bridge safety detection results; Based on several historical bridge safety inspection results, a bridge safety optimization model is constructed using a reinforcement learning algorithm, and several generated historical bridge safety optimization experiences are stored in an experience replay pool.

3. According to claim 2, a bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning, Features: The PCA method is used to reduce the dimension of the historical sensor data matrix to obtain the principal component set and the historical sensor data matrix after dimension reduction.

4. According to claim 3, a bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning, Features: The bridge cluster anomaly analysis model is constructed based on the CNN-BiLSTM algorithm, and the bridge cluster anomaly analysis model includes a matrix feature extraction module constructed based on the CNN algorithm and a bridge cluster anomaly analysis module constructed based on the BiLSTM algorithm; The bridge safety detection model is constructed based on the GSCNN-ShuffleNet algorithm, and the bridge cluster anomaly analysis model includes an image segmentation module constructed based on the GSCNN algorithm and a bridge safety detection module constructed based on the ShuffleNet algorithm; The bridge safety optimization model is constructed based on the DQN algorithm, and the bridge safety optimization model includes a deep Q network, an intelligent agent, and an experience replay pool.

5. According to claim 1, a bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning, Features: Based on the cloud data center, the real-time sensor data of several bridges are matrix-converted according to the corresponding real-time positioning data to obtain the real-time sensor data matrix of the bridge cluster, including the following steps: Based on the cloud data center, the real-time positioning data of several bridges are clustered with the bridge location as the clustering target using the FCM algorithm to obtain several real-time positioning data clusters. Extracting the real-time sensor data corresponding to each real-time positioning data in the same real-time positioning data clustering cluster, and obtaining the real-time sensor data clustering clusters of the same monitoring area; The real-time sensor data of the same monitoring area are clustered and transformed into a matrix to obtain the real-time sensor data matrix of the bridge cluster in the same monitoring area.

6. According to claim 4, a bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning, Features: Based on the cloud data center, according to the real-time sensor data matrix, the bridge cluster anomaly analysis model is used to perform bridge cluster anomaly analysis to obtain real-time bridge cluster anomaly analysis results, including the following steps: Based on the cloud data center, the real-time sensor data matrix is ​​subjected to data dimension reduction according to the principal component set to obtain the real-time sensor data matrix after dimension reduction; According to the real-time sensor data matrix, the matrix feature extraction module of the bridge cluster anomaly analysis model is used to extract matrix features and obtain real-time matrix features; According to the real-time matrix characteristics, the bridge cluster anomaly analysis module of the bridge cluster anomaly analysis model is used to perform bridge cluster anomaly analysis to obtain real-time bridge cluster anomaly analysis results.

7. According to claim 4, a bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning, Features: Based on the cloud data center, according to the real-time abnormal bridge satellite remote sensing images, the bridge safety detection model is used to perform bridge safety detection and obtain real-time bridge safety detection results, including the following steps: Based on the cloud data center, the real-time abnormal bridge satellite remote sensing image is preprocessed to obtain the preprocessed real-time abnormal bridge satellite remote sensing image; According to the real-time abnormal bridge satellite remote sensing image after preprocessing, the image segmentation module of the bridge safety detection model is used to perform image segmentation to obtain a real-time bridge segmentation image; According to the real-time bridge segmentation image, the bridge safety detection module of the bridge safety detection model is used to perform bridge safety detection to obtain real-time bridge safety detection results.

8. According to claim 4, a bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning, Features: Based on the cloud data center, according to the real-time bridge safety detection results with bridge safety risks, the bridge safety optimization model is used to optimize the bridge safety and obtain the real-time bridge safety optimization strategy, which includes the following steps: Based on the cloud data center, according to the real-time bridge safety detection results with bridge safety risks, the corresponding historical bridge safety optimization experience is retrieved in the experience playback pool to obtain several matching historical bridge safety optimization experiences; According to a number of matching historical bridge safety optimization experiences, the action space of the bridge safety optimization model is updated to obtain an updated action space; According to the real-time bridge safety detection results of the bridge with safety risks, the state space of the bridge safety optimization model is updated to obtain an updated state space; Use the updated state space as the input of the deep Q network and use the agent to generate the Q value of each possible action in the updated action space; Use the reward function to obtain the reward value of each possible action in the updated action space, and update the Q value of the possible action according to the reward value to obtain the updated Q value of each possible action; Repeat the above Q-value updating steps until the predetermined number of iterations is reached, and use a greedy strategy to select the possible action with the highest updated Q-value in each iteration as the execution action; Several execution actions are integrated to obtain a real-time bridge safety optimization strategy.

9. According to claim 1, a bridge cluster safety monitoring and early warning method based on Beidou high-precision positioning, Features: It also includes generating keys and registering identities for bridge data collection devices based on a trusted institution, obtaining corresponding public-private key pairs and signature information, returning the private key and signature information in the public-private key pairs to the corresponding bridge data collection devices, and publishing the public key in the public-private key pairs to a cloud data center; The private key is used to encrypt the real-time positioning data and real-time sensor data collected by the bridge data collection device to obtain encrypted real-time data; The signature information is used to sign the encrypted real-time data to obtain real-time signature data.

10. A bridge cluster safety monitoring and early warning system based on Beidou high-precision positioning, used to implement the bridge cluster safety monitoring and early warning method as described in any one of claims 1 to 9, Features: The system includes a cloud data center, a trusted institution, and several bridge data collection devices; The cloud data center is respectively connected to the trusted institution and a plurality of bridge data acquisition devices in communication, and the cloud data center includes a model building unit, a signature authentication and decryption unit, a matrix conversion unit, a bridge cluster abnormality analysis unit, a first safety warning unit, an image acquisition unit, a bridge safety detection unit, a second safety warning unit, a bridge safety optimization and a data sending unit which are connected in sequence; The trusted institutions are respectively connected to a plurality of bridge data collection devices for communication; The bridge data acquisition device includes a main control module, a data encryption and signature module, a Beidou high-precision positioning module, a sensor module and a data upload module. The main control module is electrically connected to the data encryption and signature module, the Beidou high-precision positioning module, the sensor module and the data upload module, and the data upload module is communicatively connected to the cloud data center and the trusted organization.

Images