An intelligent bridge inspection and evaluation system and an inspection and evaluation method

Abstract

The present invention relates to a bridge intelligent inspection and evaluation system and an inspection and evaluation method, which includes: a cluster bridge multi-source information database building module for building a cluster bridge multi-source information database according to the collected information on bridge design, construction, maintenance, diseases and service environment; an intelligent inspection device for identifying and positioning the bridge to be evaluated, taking photos of diseases, identifying the types and locations of diseases, and uploading the identified disease information to the cluster bridge multi-source information database; a cloud evaluation module for dynamically evaluating the risk of the bridge to be evaluated based on the Bayesian network using the cluster bridge multi-source information stored in the cluster bridge multi-source information database. Manual photography can greatly reduce the visual interference factors in the photos and improve the imaging quality. Moreover, the intelligent inspection device can automatically identify the types and locations of diseases, without the need for inspectors to manually enter disease information, which can improve the standardization of data entry and greatly improve the recognition accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge detection and evaluation, and in particular to an intelligent bridge inspection and evaluation system and an inspection and evaluation method. Background Art

[0002] With the development of society and economy, the demand for transportation is increasing, the number and weight of vehicles passing through are growing rapidly, and bridge defects are increasing. As bridges gradually transition from the new construction period to the maintenance period, the bridge inspection and maintenance work is increasing. Data show that when the inspection tasks increase, the number of reported bridge defect information will decrease relatively, which is extremely unfavorable for bridge operation and maintenance.

[0003] In the related technologies, in terms of intelligent inspection, some teams advocate the development of drones, inspection robots, etc. for fully automated inspection and identification. However, at present, under the premise of requiring high recognition precision and accuracy, machine learning algorithms are still not suitable for automatic identification in large fields of view and complex environments, and drones and inspection robots are still inconvenient to operate and cannot meet all environmental detection tasks. The commonly used information inspection method is to develop mobile inspection clients, but it still requires manual photography, input of bridge information, disease information, etc. Due to the different habits and work attitudes of inspection personnel, the background data description is not standardized and unified, and there are even many errors.

[0004] Therefore, it is necessary to design a new bridge intelligent inspection and evaluation system and inspection and evaluation method to overcome the above problems. Summary of the invention

[0005] The embodiments of the present invention provide an intelligent bridge inspection and evaluation system and inspection and evaluation method to solve the problems that drones and inspection robots in related technologies still have inconveniences in operation and cannot meet all environmental detection tasks. In addition, the use of mobile inspection clients will result in non-standard and inconsistent background data descriptions and even many errors due to different habits and work attitudes of inspection personnel.

[0006] In a first aspect, a bridge intelligent inspection and evaluation system is provided, which includes: a cluster bridge multi-source information database building module, the cluster bridge multi-source information database building module is used to build a cluster bridge multi-source information database based on collected bridge design, construction, maintenance, disease and service environment information; intelligent inspection equipment, the intelligent inspection equipment is used to identify and locate the bridge to be evaluated, take photos of the disease, and identify the type and location of the disease, and upload the identified disease information to the cluster bridge multi-source information database; an evaluation module, the evaluation module includes a cloud-based evaluation module, the cloud-based evaluation module uses the cluster bridge multi-source information stored in the cluster bridge multi-source information database to perform a Bayesian network-based dynamic risk evaluation of the bridge to be evaluated.

[0007] In some embodiments, the intelligent inspection device identifies the currently inspected bridge based on the global positioning system; alternatively, the intelligent inspection device automatically scans the two-dimensional code of the bridge through a camera to identify the currently inspected bridge; or, the automatic identification of the currently inspected bridge is achieved based on Bluetooth or NFC chips. After the current bridge is identified, the intelligent inspection device automatically reads multi-source information such as the basic information, structural form, historical disease information, and historical maintenance records of the inspected bridge. In some embodiments, the intelligent inspection device is provided with a shooting button. When the shooting button is clicked, the intelligent inspection device takes multiple disease photos within a preset time and records the data of the accelerometer and gyroscope when each disease photo is taken. The intelligent inspection device is further provided with a bridge disease intelligent vision recognition software module. The intelligent inspection device intelligently identifies the position of the disease in the picture and the type of the disease based on the bridge disease intelligent vision recognition software module and the taken disease photos. The intelligent inspection device is further configured to calculate the size information of the disease by using the ARKit algorithm based on the data of the accelerometer and gyroscope.

[0008] In some embodiments, the intelligent inspection device is further configured to display the type, position, and size information of the disease on the screen of the intelligent inspection device and prompt the operator to confirm whether the disease recognition is accurate. If it is accurate, the disease information is uploaded to the cluster bridge multi-source information database; otherwise, the disease information is uploaded to the cloud for manual review and correction and then uploaded to the cluster bridge multi-source information database.

[0009] In some embodiments, the bridge intelligent inspection and evaluation system further includes a disease analysis module. The disease analysis module is configured to analyze and determine whether the disease is a historical disease when the operator confirms that the disease recognition is accurate. If so, the current disease information is recorded into the matched historical disease information and uploaded to the cluster bridge multi-source information database; otherwise, a new disease is created and uploaded to the cluster bridge multi-source information database.

[0010] In some embodiments, the bridge intelligent inspection and evaluation system further includes an intelligent recognition training module. The intelligent recognition training module is configured to receive the disease information that has been manually reviewed and labeled in the cloud, incorporate it into the machine learning model for training, and then dynamically update the bridge disease intelligent vision recognition software module.

[0011] In some embodiments, the evaluation module is configured to calculate the integrity of the bridge information of the bridge to be evaluated based on the cluster bridge multi-source information stored in the cluster bridge multi-source information database, and determine whether the integrity of the bridge information meets the preset requirements. If so, it enters the Bayesian network for bridge state evaluation and gives suggestions on maintenance measures; otherwise, the evaluation result of the current bridge is corrected by a similar cluster bridge.

[0012] The formula for the integrity of bridge information is as follows:

[0013] ,

[0014] wherein, : The integrity score of the th bridge information, : The weight of the th item of bridge information, : The update time of the th item of information of the th bridge, : The actual situation of the th item of information of the th bridge, : The full score standard of the th item of information of the th bridge.

[0015] Second, a method for intelligent inspection and evaluation of bridges is provided, which includes the following steps:

[0016] Use a handheld intelligent inspection device to identify and locate the bridge to be evaluated and take photos of the diseases;

[0017] Use the intelligent visual recognition software module for bridge diseases to intelligently identify the types and locations of diseases based on the disease photos, and upload the identified disease information to the cluster bridge multi-source information database;

[0018] Evaluate the status of the bridge to be evaluated based on the Bayesian network and the bridge information stored in the cluster bridge multi-source information database.

[0019] In some embodiments, the evaluation of the status of the bridge to be evaluated based on the Bayesian network and the bridge information stored in the cluster bridge multi-source information database includes:

[0020] Calculate the integrity of the bridge information of the bridge to be evaluated based on the bridge information stored in the cluster bridge multi-source information database, and determine whether the integrity of the bridge information meets the preset requirements;

[0021] If so, use the Bayesian network to evaluate the bridge status and give suggestions on maintenance measures; otherwise, correct the integrity of the bridge information with the evaluation of the most similar cluster bridge;

[0022] The correction with the evaluation of the most similar cluster bridge includes:

[0023] Determine whether there is a bridge with the highest similarity to the bridge to be evaluated in the cluster bridge multi-source information database;

[0024] If so, correct the evaluation result of the bridge to be evaluated according to the bridge with the highest similarity in the cluster bridge multi-source information database;

[0025] Otherwise, perform Bayesian network status assessment on the bridge to be evaluated, and prompt that the bridge information is incomplete and the potential inaccuracy of the assessment.

[0026] In some embodiments, the similarity calculation formula for cluster bridges is as follows:

[0027] ,

[0028] where : Similarity score of the bridge with the most similar bridge (span) in the cluster bridge, : Similarity score between two bridges, : Take the maximum value among multiple values in the parentheses, : The th item of bridge information weight, The th bridge in the bridge cluster th item of information, : The th bridge in the bridge cluster th item of information entry time, : The th bridge in the bridge cluster; Bridge age;

[0029] When correcting:

[0030] ,

[0031] In the formula: : Final score of bridge assessment; : Completeness score of bridge information; : Similarity score of the bridge with the most similar bridge (span) in the cluster bridge; : Similar to th item of service risk assessment score (based on Bayesian network) of the bridge with the highest similarity degree; : th item of service risk assessment score (based on Bayesian network) of the th bridge; th item;

[0032] When not correcting:

[0033] ,

[0034] In the formula: : Final score of bridge assessment; : Completeness score of bridge information.

[0035] The beneficial effects brought by the technical solution provided by the present invention include:

[0036] An embodiment of the present invention provides a bridge intelligent inspection and evaluation system and an inspection and evaluation method. Since inspection personnel can take pictures of diseases with intelligent inspection equipment, manual photography can better eliminate visual interference factors and ensure the shooting quality. Moreover, the intelligent inspection equipment can automatically identify the type and location of diseases, eliminating the need for inspection personnel to manually enter disease information, which can avoid data non-standardization caused by differences in inspection personnel's habits and work attitudes. In addition, it reduces the work complexity of inspection personnel, improves work enthusiasm, and effectively reduces the misreporting rate and omission rate of inspection diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0038] Figure 1 It is a schematic structural diagram of a bridge intelligent inspection and evaluation system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0040] With the progress of informatization, the current bridge information is becoming increasingly rich. However, in bridge evaluation work, multi-source information has not been fully utilized. For example, the data of the bridge health monitoring system generally sets a single threshold as the warning line, resulting in a high false alarm rate. Currently, bridge evaluation often relies on regular inspection reports, and a large number of bridges (spans) with similar structural forms and service environments have not been analyzed associatively, resulting in a high inspection cost.

[0041] How to accurately and efficiently complete bridge inspection work and, from the perspective of cluster bridges, efficiently associate and integrate multi-source heterogeneous big data of bridges has become an important research direction.

[0042] The embodiment of the present invention provides a bridge intelligent inspection and evaluation system and an inspection and evaluation method, which can solve the problems in the related art that drones and inspection robots are still inconvenient to operate, cannot meet all environmental detection work, and using a mobile phone inspection client will result in non-standard and inconsistent background data descriptions, and even errors due to different habits and work attitudes of inspection personnel. In addition, due to the cumbersome work process, the omission rate is relatively high.

[0043] See Figure 1 As shown, a bridge intelligent inspection and evaluation system provided by an embodiment of the present invention may include: a cluster bridge multi-source information database building module, which is used to build or improve a cluster bridge multi-source information database according to the collected information on bridge design, construction, maintenance, diseases and service environment, so that the system can be applicable with or without a cluster bridge multi-source information database; an intelligent inspection device, which is used to identify and locate the bridge to be evaluated, take photos of diseases, identify the types and locations of diseases, and upload the identified disease information to the cluster bridge multi-source information database; and an evaluation module, which includes a cloud evaluation module, and the cloud evaluation module is used to perform risk dynamic evaluation based on the Bayesian network on the bridge to be evaluated based on the cluster bridge multi-source information stored in the cluster bridge multi-source information database.

[0044] In this embodiment, the intelligent inspection device can be a handheld device. Since the inspection personnel can hold the intelligent inspection device to take pictures of diseases, manual photography can ensure good lighting conditions and eliminate visual interference factors. Moreover, the intelligent inspection device can narrow the matching range according to the historical disease information of the current inspected bridge (where the inspected bridge is also the bridge to be evaluated), improve the recognition accuracy, and the intelligent inspection device can automatically identify the types and locations of diseases without the need for inspection personnel to manually enter disease information, reducing the inspection workload, and can largely avoid problems such as data omission, misreporting, and non-standardization caused by different habits and work attitudes of inspection personnel, and further improve the accuracy of inspection data reporting.

[0045] Furthermore, in the absence of a multi-source information database for cluster bridges, during the preliminary inspection process, the inspection personnel and cloud (backend) staff can jointly improve the bridge information and build a multi-source information database for cluster bridges. Specifically, first, the cloud (backend) staff collect bridge information and establish a bridge file (structural form, bridge age, span, BCI level, etc.). Then, during the inspection process, the bridge database is improved. When the inspection personnel conduct the inspection, they match the bridge name, create new diseases and take photos (indicating the disease size and specific location), and upload the inspected bridge name and disease information to the cloud. Then, the backend staff perform manual processing, mainly including manually annotating diseases in the backend (classifying the diseases and marking the disease areas in the pictures for machine learning training), annotating bridge information (including bridge design, construction data, regular bridge inspections, etc.), and associating with the bridge monitoring system (if any). During the database construction period, the main work is completed by the cloud staff, and the workload of frontline inspection and maintenance personnel remains basically unchanged. Finally, with the progress of the daily inspection and maintenance work of urban cluster bridges, the construction and improvement of the multi-source information database for cluster bridges are gradually completed.

[0046] Preferably, the intelligent inspection device can be based on a mature smartphone platform and used in conjunction with an intelligent inspection APP. The intelligent inspection device can have a camera with a pixel count of over 24 million, a high-precision accelerometer and gyroscope, support 5G communication, and can also expand supplementary lighting devices and laser ranging devices through a data interface. The expansion devices are attached closely behind the device for the convenience of inspection personnel.

[0047] In some embodiments, when identifying and positioning the bridge to be evaluated, an intelligent bridge positioning method can be adopted. The main methods can be: (1) The intelligent inspection device assists in identifying the current inspected bridge based on the geographical positioning system and automatically reads the current inspected bridge information; for example, through the mobile phone GNSS positioning system, the current inspected bridge is determined, and the bridge information is automatically read, narrowing the bridge disease range and improving the accuracy of AI visual recognition. Or, (2) The intelligent inspection device can automatically scan the QR code of the current inspected bridge based on the camera for identification and positioning; that is, QR codes are posted on the bridge guardrail, bottom of the beam, bridge pier, etc. Before clicking to take a photo of the disease, the camera is in the on state, and it can automatically and real-time identify the nearby posted QR code to accurately determine the current inspected bridge (span) and the precise inspection location. Or, (3) Based on the Bluetooth and NFC chip methods, the automatic identification of the current inspected bridge is realized to determine the inspection location. After completing the identification of the current bridge, the basic information, structural form, historical disease information, historical maintenance records, and other multi-source information of the inspected bridge are automatically read. Of course, in other embodiments, the current inspected bridge can also be determined by manual input. For areas where the GNSS system may have positioning deviations or there is no GPS signal, manual selection can be adopted.

[0048] In some alternative embodiments, the intelligent inspection device may be provided with a shooting button. When the inspector clicks the shooting button, the intelligent inspection device can take multiple disease photos within a preset time and record the data of the accelerometer and gyroscope when each disease photo is taken. For example, it can automatically and continuously take multiple photos within the time of 1 - 2 s. There is also a bridge disease intelligent vision recognition software module in the APP of the intelligent inspection device. Among them, this model can be a bridge disease intelligent vision recognition software module based on machine vision, enabling the intelligent inspection device to intelligently identify the position of the disease in the picture and the type of the disease based on the bridge disease intelligent vision recognition software module and the taken disease photos. The APP of the intelligent inspection device can also calculate the size information of the disease based on the data of the accelerometer, gyroscope, and the ARKit algorithm.

[0049] In this embodiment, the ARKit algorithm realizes the visual ranging principle as follows: ARKit is a Visual Inertial Odometry (VIO) system with the ability of simple 2D plane detection. The VIO system tracks the moving distance of the device in 6D space. 6D represents the xyz movement (translation) in the 3D world, plus the 3D movement (rotation) of pitch / yaw / roll. The VIO technology can track the position of the device in space in real time through software. The user's pose can be tracked through the visual (camera) system. For this purpose, it is necessary to match the points in the real world with a pixel on each frame of the image captured by the camera sensor. In addition, the user's pose needs to be tracked through the inertial system (accelerometer and gyroscope). Subsequently, the output results of these systems will be merged through the Kalman Filter to determine which system can provide the best estimate of the user's "actual" position, and the position update will be published through the ARKit SDK. The distance between two points in the figure can be calculated from the coordinates given by the ARKit algorithm.

[0050] In some embodiments, the intelligent inspection device is further configured to display identification information such as the type, location, and size of the disease on the screen of the intelligent inspection device, and prompt manual confirmation of whether the disease identification is accurate; if accurate, upload the disease information to the cluster bridge multi-source information database; otherwise, upload the disease information to the cloud for manual review and correction, and then upload it to the cluster bridge multi-source information database. In this embodiment, when confirming the disease identification, the intelligent inspection device can directly display the disease identification information on the screen for the inspection personnel to verify. The inspection personnel only need to click correct or incorrect. When the identification result is inaccurate, the disease picture, accelerometer information, gyroscope data, visual solution result, and disease area marking result are uploaded to the cloud server, and the background staff will classify and mark the disease area.

[0051] In some alternative embodiments, the bridge intelligent inspection and evaluation system may further include a disease analysis module. The disease analysis module is configured to analyze and determine whether the disease is a historical disease when manual confirmation of the disease identification is accurate; if so, record the current disease information into the matched historical disease information and upload it to the cluster bridge multi-source information database; otherwise, create a new disease and upload it to the cluster bridge multi-source information database. In this embodiment, when the disease identification result is accurate (directly identified accurately or accurately identified after background correction), the disease is matched with the historical diseases of the bridge (span). If it is a new disease, a new disease record is established, including the disease location, inspection time, disease type, disease photo, disease area, disease size, etc. If there is a matched historical disease for the bridge (span), add the daily disease information to this record and generate a disease development record; synchronize the newly added or updated disease record to the cluster bridge multi-source information database.

[0052] In this embodiment, the system can automatically identify diseases and compare them with the historical disease database of the bridge (span) to be evaluated, effectively avoiding the problem of duplicate recording of a disease in the traditional management and maintenance platform. After successfully matching with the historical diseases, the disease information will be integrated to form a disease time-series evolution report for the maintenance personnel to refer to. And records such as disease records, health monitoring, and regular inspections can be transmitted back to the cluster bridge multi-source information database in real time.

[0053] Preferably, the intelligent bridge inspection and evaluation system may further include an intelligent recognition training module, which can be set in the cloud for disease recognition training. The intelligent recognition training module is used to receive the disease information that has been manually reviewed and annotated in the cloud, incorporate it into the machine learning model for training, and then dynamically update the intelligent visual recognition software module for bridge diseases. That is, after manual correction, the correct results are incorporated into the machine learning model for training, and the intelligent visual recognition software module for bridge diseases is dynamically updated until the recognition result is accurate. In this embodiment, by setting up the intelligent recognition training module for machine learning training, the image recognition model for bridge diseases can be dynamically updated, and the recognition results can be feedback corrected, improving the recognition accuracy.

[0054] In some embodiments, the evaluation module can be used to calculate the integrity of the bridge information of the bridge to be evaluated based on the multi-source information of the cluster bridges stored in the multi-source information database of the cluster bridges, and determine whether the integrity of the bridge information meets the preset requirements; if so, enter the Bayesian network for bridge status evaluation and give suggestions on maintenance measures; otherwise, correct the evaluation results of the current bridge with the most similar bridge in the cluster bridges. Among them, when a certain inspection bridge uploads the last inspection information for a period of time, for example, 15 minutes after the last information of a certain bridge is uploaded, the bridge is evaluated. The evaluation module can be divided into five parts: calculation of bridge information integrity, Bayesian network evaluation of bridge status, similarity matching of cluster bridges, correction of cluster bridge evaluation, suggestions on maintenance measures and dynamic update.

[0055] (1) Regarding the calculation of bridge information integrity. The evaluation module can have an evaluation system for evaluating the integrity of bridge information.

[0056] Users can set the integrity of bridge information and various weights according to the situation, which generally includes: regular bridge inspection reports, bridge load test data, bridge inspection results, bridge design information, bridge construction information, bridge traffic flow information, heavy vehicle passing statistics information of the bridge, etc. The formula for the integrity of bridge information is as follows:

[0057] ,

[0058] Among them, : The integrity score of the th bridge information, : The weight of the th item of bridge information, : The update time of the th item of the th bridge, : The actual situation of the th item of the th bridge, : The The full - score standard for the first piece of information of a bridge.

[0059] (2) Bayesian network assessment of bridge status.

[0060] In this embodiment, for example, when the calculated integrity of bridge information ≥ 90 points, the Bayesian network can be directly used for assessment. When the calculated integrity of bridge information < 90 points, the bridge information in the cluster bridge library can be retrieved for cluster bridge similarity matching.

[0061] (3) Regarding cluster bridge similarity matching.

[0062] The most similar bridge (span) can be queried in real - time from the multi - source information database of cluster bridges. The cluster bridge similarity calculation formula is as follows:

[0063] ,

[0064] Where, : The similarity score between the bridge and the most similar bridge (span) in the cluster bridges, : The similarity score between two bridges, : Take the maximum value among multiple values in the brackets, : The weight of the th piece of bridge information, The th piece of information of the th bridge in the bridge cluster, The th piece of information input time of the th bridge in the bridge cluster, Bridge age of the

[0065] : Define the operation symbol 1, Operation: The similarity comparison operation of the same - type information of two bridges. The rules can be set according to the actual situation of each structural - form bridge in the cluster bridges. Four common scoring methods:

[0066] ① Piece - wise scoring:

[0067] The span of Bridge A is a, and the span of Bridge B is b (a ≥ b > 0).

[0068] Then The operation score is shown in Table 1 below:

[0069] Table 1: Span Operation score table

[0070]

[0071] ② Formula scoring:

[0072] For example, for bridge traffic flow: a vehicles per lane per hour on Bridge A and b vehicles per lane per hour on Bridge B, then The calculated score is: .

[0073] ③ 01 scoring:

[0074] For example, bridge structural forms: hollow slab bridges, T-beam bridges, concrete box girder bridges, steel box girder bridges, composite beam bridges, variable cross-section box girder bridges, T-frame bridges, cable-stayed bridges, suspension bridges, arch bridges, etc.

[0075] When the structural forms are the same The calculated score is: 100 points; when they are different The calculated score is: 0 points.

[0076] ④ Composite scoring: On the basis of sectional scoring, the 01 scoring or formula scoring method can be adopted.

[0077] (3) Cluster bridge evaluation and correction. The correction plan can be combined according to the calculated bridge information integrity and the value of the cluster bridge similarity. For bridges with incomplete bridge information (that is, not meeting the preset integrity requirements), the following judgment results as shown in Table 2 are given after evaluation:

[0078] Table 2: Judgment Table for Cluster Bridge Correction Conditions

[0079]

[0080] The score calculation formula is as follows:

[0081] When correcting:

[0082] ,

[0083] In the formula: : The final score of bridge evaluation; : The score of bridge information integrity; : The similarity score between the bridge and the most similar bridge (span) in the cluster bridge; : Compared with The service risk assessment score of the th item of the bridge with the highest similarity degree to the bridge (based on the Bayesian network); : The service risk assessment score of the th item of the bridge (based on the Bayesian network).

[0084] When not correcting:

[0085] ,

[0086] In the formula: : The final score of bridge assessment; : The score of bridge information integrity.

[0087] (4) Maintenance measure suggestions and dynamic update.

[0088] Based on the conclusion of condition assessment, the maintenance methods are given as shown in Table 3, and the specific content is given by the system in combination with the "Technical Standard for Urban Bridge Maintenance" (CJJ99-2017).

[0089] Table 3 Suggestions for maintenance measures

[0090]

[0091] After correcting and giving the score, the multi-source information database of cluster bridges is fed back in real time, and the assessment information of this bridge is updated with the score. The system gives suggestions and early warnings according to the table of the final score of cluster bridge Bayesian network assessment - maintenance and alarm measures.

[0092] The corrected content will be transmitted to the cloud server in real time. The server will train regularly (usually once a day) in combination with the feedback data, and dynamically update the training results to the intelligent terminal to dynamically improve the recognition accuracy.

[0093] In this embodiment, the edge computing terminal stores all the basic data of the bridges on the previous day. When updating the situation of a certain bridge, Bayesian network evaluation and cluster correction are performed according to the data of this bridge (span) today and the database of other bridges on the previous day, and the preliminary evaluation score is given on the spot; after this bridge is updated, it is uploaded to the cloud server, and the cloud server will perform dynamic evaluation of the cluster bridges according to the real-time data, and give the evaluation score, risk items and rectification suggestions, where the rectification suggestions can be given in combination with the current national standard "Technical Standard for Urban Bridge Maintenance" (CJJ99-2017). And many methods and examples are given, which can be selected by different cluster bridges according to the project and combined with the project, and have good applicability.

[0094] The embodiment of the present invention also provides a bridge intelligent inspection and evaluation method. The bridge intelligent inspection and evaluation method in the embodiment of the present invention can be implemented by using the bridge intelligent inspection and evaluation system provided in any of the above embodiments. Among them, the bridge intelligent inspection and evaluation method may include the following steps:

[0095] Step 1: Use a handheld intelligent inspection device to identify and locate the bridge to be evaluated and take photos of the diseases.

[0096] Step 2: Use the intelligent vision recognition software module for bridge diseases to intelligently identify the types and locations of diseases based on the disease photos, and upload the identified disease information to the cluster bridge multi-source information database.

[0097] Step 3: Evaluate the status of the bridge to be evaluated based on the Bayesian network and the bridge information stored in the cluster bridge multi-source information database.

[0098] Further, the evaluation of the status of the bridge to be evaluated based on the Bayesian network and the bridge information stored in the cluster bridge multi-source information database may include: calculating the integrity of the bridge information of the bridge to be evaluated based on the bridge information stored in the cluster bridge multi-source information database, and determining whether the integrity of the bridge information meets the preset requirements; if so, enter the Bayesian network for bridge status evaluation and give suggestions on maintenance measures; otherwise, correct the evaluation results of the bridge to be evaluated with the evaluation results of similar cluster bridges.

[0099] In some embodiments, the correction of the evaluation results of the bridge to be evaluated with the evaluation results of similar cluster bridges may include: determining whether there is a bridge in the cluster bridge multi-source information database that matches the similarity of the bridge to be evaluated; if so, correct the evaluation score of the bridge to be evaluated according to the evaluation results of the bridge that matches the similarity in the cluster bridge multi-source information database, otherwise, perform Bayesian network status evaluation on the bridge to be evaluated and prompt that the bridge information is incomplete and the evaluation may be potentially inaccurate.

[0100] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by terms such as "upper" and "lower" is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. Unless otherwise clearly specified and limited, the terms "installation", "connection" and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0101] It should be noted that in the present invention, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising said element.

[0102] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather will be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A bridge intelligent inspection and evaluation system, characterized in that: It includes: A cluster bridge multi-source information database building module, wherein the cluster bridge multi-source information database building module is used to build a cluster bridge multi-source information database based on the collected bridge design, construction, maintenance, disease and service environment information; Intelligent inspection equipment, which is used to identify and locate the bridge to be evaluated, take photos of the defects, identify the type and location of the defects, and upload the identified defect information to the cluster bridge multi-source information database; An evaluation module, the evaluation module comprising a cloud evaluation module, the cloud evaluation module using the cluster bridge multi-source information stored in the cluster bridge multi-source information database to perform a Bayesian network-based dynamic risk evaluation on the bridge to be evaluated; The evaluation module is used to calculate the bridge information integrity of the bridge to be evaluated based on the cluster bridge multi-source information stored in the cluster bridge multi-source information database, and determine whether the bridge information integrity meets the preset requirements; if so, enter the Bayesian network to evaluate the bridge status and give maintenance measures; otherwise, correct the current bridge evaluation result by using similar cluster bridges; The formula for bridge information completeness is as follows: ， in, : No. Bridge information completeness score, : No. The bridge information weight, : No. Bridge No. Item information update time, : No. Bridge No. The actual situation of the information, : No. Bridge No. Full score criteria for item information; The present bridge assessment results are corrected by using similar cluster bridges, including: Determine whether there is a bridge in the cluster bridge multi-source information database that has the highest similarity to the bridge to be evaluated; If so, the evaluation result of the bridge to be evaluated is modified according to the bridge with the highest similarity in the multi-source information database of the cluster bridges; Otherwise, a Bayesian network status assessment is performed on the bridge to be assessed, and it is indicated that the bridge information is incomplete and the assessment is potentially inaccurate; The cluster bridge similarity calculation formula is as follows: ， in, : The similarity score between the bridge and the most similar bridge in the cluster, : Similarity score of two bridges, : Take the maximum value among the multiple values ​​in brackets. : No. The bridge information weight, : The bridge cluster Bridge No. Item information, : The bridge cluster Bridge No. Item information, : Comparison operation of similarity between two bridges of the same type, : The bridge cluster Bridge No. Time when the information is entered. : The bridge cluster Bridge No. Time when the information is entered. : The bridge cluster The age of the bridge, : The bridge cluster Age of bridges; When corrected: ， Where: : Final score of bridge assessment; : Bridge information completeness score; : Similarity score between the bridge and the most similar bridge in the cluster; :and The bridge with the highest degree of similarity Service risk assessment score; : Bridge Service risk assessment score; When not corrected: ， Where: : Final score of bridge assessment; : Bridge information completeness score.

2. The intelligent bridge inspection and evaluation system according to claim 1, characterized in that: The intelligent inspection device assists in identifying the current inspection bridge based on the geographic positioning system; or, the intelligent inspection device automatically scans the QR code of the bridge based on the camera to realize the identification of the current inspection bridge; or, realizes automatic identification of the current inspection bridge based on Bluetooth or NFC chip; after completing the identification of the current bridge, automatically reads the basic information, structural form, historical disease information, historical maintenance records and other multi-source information of the inspection bridge.

3. The intelligent bridge inspection and evaluation system according to claim 1, characterized in that: The intelligent inspection device is provided with a shooting button. When the shooting button is clicked, the intelligent inspection device takes a plurality of disease photos within a preset time, and records the data of the accelerometer and the gyroscope when each disease photo is taken; The intelligent inspection equipment is also provided with a bridge disease intelligent visual recognition software module, and the intelligent inspection equipment intelligently recognizes the location of the disease in the picture and the type of the disease based on the bridge disease intelligent visual recognition software module and the taken disease photos; The intelligent inspection device is also used to calculate the size information of the disease based on the data of the accelerometer and the gyroscope using the ARKit algorithm.

4. The intelligent bridge inspection and evaluation system according to claim 3, characterized in that: The intelligent inspection device is also used to display the type, location and size information of the disease on the screen of the intelligent inspection device, and prompt manual confirmation whether the disease identification is accurate; If accurate, the disease information is uploaded to the cluster bridge multi-source information database; Otherwise, the disease information is uploaded to the cloud for manual review and correction before being uploaded to the cluster bridge multi-source information database.

5. The intelligent bridge inspection and evaluation system according to claim 4, characterized in that: The intelligent bridge inspection and evaluation system also includes a defect analysis module, which is used to analyze and determine whether the defect is a historical defect when the defect identification is manually confirmed to be accurate; if so, the current defect information is recorded in the matched historical defect information and uploaded to the cluster bridge multi-source information database; otherwise, a new defect is created and uploaded to the cluster bridge multi-source information database.

6. The intelligent bridge inspection and evaluation system according to claim 4, characterized in that: The bridge intelligent inspection and evaluation system also includes an intelligent recognition training module, which is used to receive the disease information that has been manually reviewed and labeled in the cloud, incorporate it into the machine learning model for training, and then dynamically update the bridge disease intelligent visual recognition software module.

7. A bridge intelligent inspection and evaluation method, characterized in that: It includes the following steps: Handheld intelligent inspection equipment can identify and locate the bridges to be assessed and take photos of the defects; Based on the disease photos, the bridge disease intelligent visual recognition software module is used to intelligently identify the type and location of the disease, and the identified disease information is uploaded to the cluster bridge multi-source information database; Evaluate the status of the bridge to be evaluated based on the Bayesian network and the bridge information stored in the cluster bridge multi-source information database; The evaluating the state of the bridge to be evaluated based on the Bayesian network and the bridge information stored in the cluster bridge multi-source information database includes: Calculating the bridge information integrity of the bridge to be evaluated based on the bridge information stored in the cluster bridge multi-source information database, and judging whether the bridge information integrity meets the preset requirements; If yes, then use Bayesian network to evaluate the bridge status and give maintenance measures; otherwise, evaluate the most similar cluster bridges and correct the bridge information completeness; The formula for bridge information completeness is as follows: ， in, : No. Bridge information completeness score, : No. The bridge information weight, : No. Bridge No. Item information update time, : No. Bridge No. The actual situation of the information, : No. Bridge No. Full score criteria for item information; The most similar cluster bridge assessment revisions include: Determine whether there is a bridge in the cluster bridge multi-source information database that has the highest similarity to the bridge to be evaluated; If so, the evaluation result of the bridge to be evaluated is modified according to the bridge with the highest similarity in the multi-source information database of the cluster bridges; Otherwise, a Bayesian network status assessment is performed on the bridge to be assessed, and it is indicated that the bridge information is incomplete and the assessment is potentially inaccurate; The cluster bridge similarity calculation formula is as follows: ， in, : The similarity score between the bridge and the most similar bridge in the cluster, : Similarity score of two bridges, : Take the maximum value among the multiple values ​​in brackets. : No. The bridge information weight, : The bridge cluster Bridge No. Item information, : The bridge cluster Bridge No. Item information, : Comparison operation of similarity between two bridges of the same type, : The bridge cluster Bridge No. Time when the information is entered. : The bridge cluster Bridge No. Time when the information is entered. : The bridge cluster The age of the bridge, : The bridge cluster Age of bridges; When corrected: ， Where: : Final score of bridge assessment; : Bridge information completeness score; : Similarity score between the bridge and the most similar bridge in the cluster; :and The bridge with the highest degree of similarity Service risk assessment score; : Bridge Service risk assessment score; When not corrected: ， Where: : Final score of bridge assessment; : Bridge information completeness score.

Images