Engineering structure damage identification method based on dynamic response sensitive component and light-weight monitoring system

Abstract

The application discloses an engineering structure damage identification method and a light-weight monitoring system based on a dynamic response sensitive component, comprising a structure vibration sensing module, a high-performance micro control module, a damage identification algorithm module, a cloud storage module, a structure damage early warning module and a user terminal module; the high-performance micro control module is connected with the user terminal module through the cloud storage module; the high-performance micro control module is also connected with the structure damage early warning module through the damage identification algorithm module, the structure damage early warning module is connected with the cloud storage module, and the structure vibration sensing module serves as an input module of the high-performance micro control module; through response sensitive component selection based on energy intensity information entropy, an energy intensity damage index sensitive to local early micro damage and identification are developed. The application significantly improves the accuracy and efficiency of engineering structure damage identification, and the method and device have low power consumption, high cost performance and wide applicability, and have important engineering application value and significance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of structural health monitoring and damage identification in infrastructure and civil engineering, and particularly relates to a structural damage identification method based on dynamic response sensitive components and a lightweight structural damage monitoring system with edge computing function. BACKGROUND

[0002] Infrastructure such as high-rise buildings, large venues and industrial structures are subjected to environmental and various load effects during service, and the probability of structural material aging, corrosion fatigue or other structural damage gradually increases. Once damage or destruction occurs, it will cause serious consequences and pose a serious threat to public safety. In order to prevent the occurrence of serious destructive damage, the main load-bearing floors and components of engineering structures should be monitored for early, local and minor damage. When internal damage occurs or is blocked, it is impossible to directly measure the structural damage, and such situations often occur in practice. Therefore, it is of great theoretical value and practical significance to study flexible and accurate engineering structural damage identification methods and develop high-performance, lightweight structural damage monitoring devices.

[0003] Traditional structural health monitoring systems mainly use sensors, data acquisition systems and analysis systems installed on buildings, bridges and other infrastructure to monitor their status and performance in real time. These systems can detect important parameters such as structural deformation, stress and dynamic response, and are used to identify potential safety hazards in a timely manner to ensure the safety and reliability of the structure. However, there are still some challenges in its application. The cost of sensors and data acquisition equipment of the current structural health monitoring system is relatively high, and the amount of data generated during the monitoring process is huge, which requires a large amount of computing and network resources to process and store the data, limiting its widespread application.

[0004] Damage identification is one of the primary tasks of structural health monitoring systems and a crucial means of ensuring structural safety. Over the past few decades, methods for indirectly determining structural health by measuring the dynamic response of a structure have attracted considerable attention. While these methods offer the advantages of easy measurement and global diagnostic capabilities, they also face the challenge of accurately extracting structural damage information from the dynamic response. Current approaches to damage identification based on structural vibration include modal parameter methods, modern signal analysis, physical parameter identification, statistical pattern recognition, and artificial intelligence. Most of these methods involve the concepts of dynamic response energy or signal energy. Energy-based damage indicators rely on accurate modal shape calculations, resulting in high computational and engineering costs. They are insensitive to minor structural damage, and the accuracy of damage location and severity identification requires further improvement. This is because the structural response signal contains a large number of components that are insensitive to structural damage. These components are primarily susceptible to factors such as excitation randomness, uncertainty, and noise. Therefore, effectively selecting components that are sensitive to structural damage is a key approach to improving damage identification accuracy. Current methods for selecting sensitive components generally require prior knowledge of the structural damage, which is difficult to achieve for practical engineering structures. Summary of the Invention

[0005] Traditional structural health monitoring systems suffer from numerous challenges, including high acquisition equipment costs, massive amounts of monitored data, and significant consumption of computing and network resources. Furthermore, the damage identification methods used in these monitoring systems rely on precise modal shape solutions, resulting in high computational and engineering costs. Furthermore, these systems are influenced by non-sensitive components of the dynamic response, making damage indicators insensitive to minor structural damage and providing low accuracy in identifying damage locations.

[0006] In response to the above-mentioned deficiencies in traditional structural health monitoring systems and damage identification methods, the present invention provides an engineering structure damage identification method and a lightweight monitoring system based on dynamic response sensitive components. By proposing a new sensitive component selection method, the dynamic response sensitive component is selected to establish an energy damage index that is sensitive to local early minor damage, which can greatly improve the accuracy of engineering structure damage identification. By using edge computing technology, the damage identification method is deployed in the monitoring system, which can analyze and process the structural dynamic response in real time near the data source, and perform diagnosis and early warning of structural damage, thereby building a structural monitoring system that integrates data acquisition, calculation, transmission and early warning. The invention has low power consumption, high cost performance and wide applicability, and has good engineering application prospects and significance.

[0007] In order to achieve the above-mentioned purpose, the engineering structure damage identification method based on sensitive components of dynamic response and the light monitoring system provided by the application are characterized in that the light monitoring system mainly comprises: an acceleration sensor (1), a core processor (2), a signal acquisition unit (3), a data preprocessing unit (4), a database unit (5), a response signal decomposition unit (6), a sensitive component selection unit (7), a sensitive component reconstruction unit (8), a damage index calculation unit (9), a early warning response unit (10), a remote communication early warning (11), a data storage playback unit (12), a data visualization unit (13), a user mobile terminal (14), and a user computer terminal (15).

[0008] The light monitoring system comprises six main functional modules, specifically: 1, a structure vibration sensing module: mainly comprising an acceleration sensor as a sensing element, used for measuring the acceleration response or displacement response of the key monitoring position and the key measuring point of the structure facility; 2, a high-performance micro control module: mainly comprising a core processor (2) and an embedded signal acquisition unit (3), a data preprocessing unit (4), and a database unit (5), used for collecting and transmitting the structure dynamic response to other core modules, and bearing the edge computing function; 3, a damage identification algorithm module: disposed on the high-performance micro control module, comprising a response signal decomposition unit (6), a sensitive component selection unit (7), a sensitive component reconstruction unit (8), and a damage index calculation unit (9), used for analyzing and processing the low-level original response signal, obtaining high-level feature data that can evaluate the structure damage condition, and making a judgment on the structure damage; 4, a cloud storage module: comprising a data storage playback unit (12) and a data visualization unit (13), used for long-term storing the structure vibration response data and all high-level feature data of the key time nodes and the abnormal time nodes, and can be called and visually displayed; 5, a structure damage early warning module: comprising a early warning response unit (10) and a remote communication early warning unit (11), used for receiving the high-level feature data of the damage identification algorithm module and determining whether to perform damage early warning, when the structure damage is identified, the early warning response unit (10) sends an on-site early warning, and the remote communication early warning unit (11) sends the early warning information to the user terminal; 6, a user terminal module: divided into a user mobile terminal (14) and a user computer terminal (15), used for receiving the early warning information and querying and accessing the historical monitoring data of the structure;

[0009] As a further technical scheme of the application, the structure vibration sensing module is characterized in that a MEMS accelerometer is mainly used as the main acceleration sensor (1).

[0010] As a further technical scheme of the present application, the high-performance micro control module is characterized in that the signal acquisition unit (3) is connected with the structural vibration sensing module to realize the function of dynamic response data acquisition; the data preprocessing unit (4) analyzes and processes the output signal of the signal acquisition unit (3) to obtain stable and high-quality structural dynamic response data, which are transmitted to the database unit (5) and the damage identification algorithm module, respectively; the database unit (5) is used for temporarily storing the structural dynamic response data.

[0011] As a further technical scheme of the present application, the signal acquisition unit (3) is characterized in that ZigBee technology is used to realize wireless network transmission, the sampling frequency and single sampling duration of the acceleration sensor (1) are determined according to the monitored structural dynamic characteristics, the sampling frequency can be set to a frequency value greater than 2 times the fifth-order natural frequency value of the monitored structure, the sampling duration can be set to 600s-3600s, and the interval between two response sampling processes is set to 5s-10s; the data preprocessing unit (4) performs zero-mean processing and detrend processing on the output signal of the signal acquisition unit (3); and the database unit (5) mainly includes a database for storing short-term monitoring data.

[0012] As a further technical scheme of the present application, the damage identification algorithm module is characterized in that the structural damage identification method based on dynamic response sensitive components proposed in the present application is packaged as a program and deployed on the core processor (2) to realize the function of real-time calculation and analysis of structural dynamic response and identification of structural damage condition at the edge of the high-performance micro control module.

[0013] As a further technical scheme of the present application, the cloud storage module is characterized in that data can be uploaded to the cloud storage module through application programming interface (API), web page or client, etc., and the user end module can manage the data stored in the cloud storage server through API, web page, etc., such as online browsing, uploading, downloading, deleting, renaming, sharing, etc.

[0014] As a further technical scheme of the present application, the structural damage early warning module is characterized in that the early warning response unit (10) mainly includes two kinds of hardware, i.e., high-brightness warning light and high-response whistle device, which are externally connected with the core processor (2) and receive the start instruction of the damage identification algorithm module to issue an alarm including light and whistle on site. The remote communication early warning unit (11) mainly includes a 5G communication module, which is externally connected with the core processor (2) and receives the start instruction of the damage identification algorithm module to send early warning information to the user mobile device end.

[0015] As a further technical scheme of the present application, the lightweight monitoring system is characterized in that the main construction steps include: step 1: for the actual structure monitoring object, determine the key monitoring position and key measuring point, and establish a structure vibration sensing module; step 2: design and develop a high-performance micro control module, establish a signal acquisition unit (3), a data preprocessing unit (4), and a database unit (5) according to the functional requirements; connect the signal acquisition unit (3) with the structure vibration sensing module, and build the data transmission path among the signal acquisition unit (3), the data preprocessing unit (4), and the database unit (5); step 3: encapsulate the structure damage identification method based on dynamic response sensitive components into a program, deploy it on the high-performance micro control module, and establish a data transmission path between the data preprocessing unit (4) and the database unit (5); step 4: create a cloud storage module to realize the function of uploading data from the database unit (5) and the damage identification algorithm module to it, store the dynamic response data of the key time nodes and abnormal time nodes, and the damage index result data; step 5: design and expand the structure damage early warning module; step 6: build a user end for receiving early warning information and querying historical monitoring data of the structure.

[0016] As a further technical scheme of the present application, the engineering structure damage identification method based on dynamic response sensitive components is characterized in that the specific steps are as follows:

[0017] Step 3-1: according to the structure characteristics and analysis requirements, arrange acceleration sensors (1) near the monitoring position, the signal acquisition unit (3) acquires the structure dynamic response under healthy state as the baseline data, and acquires the dynamic response data of each measuring point as the measured data in real time according to the set sampling time length;

[0018] Step 3-2: select a wavelet base function and determine the decomposition level, the wavelet base function can be selected as discrete Meyer function and Daubechies (dbN) function, the decomposition layer j can be set to 3-8 layers, the baseline data and the measured data are respectively decomposed by wavelet packet according to formula (1), and then the last layer components are reconstructed according to formula (2); the vibration response data of all measuring points are decomposed and reconstructed into i wavelet packet components, denoted as According to formula (3), the energy values of all wavelet packet components are calculated, and according to formula (4), the energy matrices E j,h and E j,d of the wavelet packet components of the baseline data and the measured data are respectively constructed;

[0019]

[0020] In the formula, are the wavelet packet coefficients of the adjacent layer nodes after the response is decomposed by the wavelet packet, h and g are low-pass and high-pass filters respectively, the subscript i represents the wavelet packet component number, the subscript k represents the wavelet packet coefficient number, and l is the number of processed sequence data points.

[0021]

[0022] Where, E n i Represents the energy value of the i-th wavelet packet component of the monitoring signal of the n-th measuring point on the structure. Similarly, the energy matrix E j The other elements in the file are named according to this rule.

[0023] Step 3-3: From the energy matrix E of the wavelet packet components j,h and E j,d Extract and construct the energy vector E of the same component of all measurement points h and E d , and calculate the energy intensity vector D of any wavelet packet component according to formula (5), and then calculate the energy intensity information entropy value H of any wavelet packet component according to formulas (6) and (7) wpt (b), finally, select the energy intensity information entropy value H in the wavelet packet component wpt (b) The smallest first 15% to 30% components are used as vibration response sensitive components;

[0024] D = log 10 (E d / E h ) (5)

[0025]

[0026] Where, P b,c It represents the ratio of the energy intensity value of the bth wavelet packet component of the monitoring signal of the cth measuring point of the structure to the total energy intensity value of the bth wavelet packet component of the monitoring signal of all measuring points of the structure, D c b and D q b They represent the energy intensity values ​​of the bth wavelet packet component of the monitoring signal at the cth and qth measuring points of the structure respectively.

[0027] Step 3-4: According to formula (2), the vibration response sensitive components selected in step 3-3 are reconstructed by wavelet packets, and the non-sensitive component information is eliminated to obtain the reconstructed signal data a of the reference data and the data to be measured. h (t) and a d (t);

[0028] Step 3-5: Calculate the sensitive component energy value E of the reconstructed signal data of each measuring point in the healthy state and the test state according to formula (8):r,h and E r,d According to formula (9), the energy intensity of the vibration response sensitive component of the to-be-tested structure is calculated as a damage index, and the damage condition of the to-be-tested structure can be directly analyzed.

[0029]

[0030] S E = log 10 (E r,d / E r,h ) (9)

[0031] In the formula, E r represents the energy value of the reconstructed signal of the sensitive component of the monitoring signal of each measuring point, and S E represents the energy intensity value of the sensitive component in the monitoring signal of each measuring point of the to-be-tested structure.

[0032] Compared with the prior art, the advantages of the present application are as follows:

[0033] 1) The dynamic performance parameters of the structure and facility can be automatically and all-weather monitored, and the tedious monitoring and collection work of the staff and the human error caused by the operation error are avoided; at the same time, the overall cost of the developed lightweight monitoring system is relatively low, and the system has low power consumption, high cost performance and wide applicability.

[0034] 2) The edge computing technology is adopted, the damage identification algorithm is embedded on the core processor (2), the structure dynamic response data can be directly analyzed and processed, the low-level massive monitoring data is converted into high-level feature-level damage index, the real-time transmission is facilitated, the consumption of data transmission is reduced, the congestion caused by insufficient network width is avoided, and the efficiency of structure monitoring and damage diagnosis is greatly improved, which has important engineering significance.

[0035] 3) According to the engineering requirement of extracting the damage information of the structure from the vibration response, the traditional vibration-based method is improved, the early, local and small damage identification of the structure is not accurate, and the structure damage identification method based on the dynamic response sensitive component is provided, and the local small damage at the key position can be accurately identified. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 is the composition diagram of the engineering structure damage identification method and the lightweight monitoring system based on the dynamic response sensitive component provided by the embodiment of the present application;

[0037] Figure 2 is the specific implementation flowchart of the engineering structure damage identification method and the lightweight monitoring system based on the dynamic response sensitive component provided by the embodiment of the present application;

[0038] Figure 3is a flow chart of the engineering structure damage identification method based on the dynamic response sensitive component provided by the embodiment of the application;

[0039] Figure 4 is a building structure schematic diagram and a MEMS accelerometer measuring point arrangement schematic diagram provided by the embodiment of the application;

[0040] Figure 5 is a wavelet packet component energy intensity information entropy result diagram provided by the embodiment of the application;

[0041] Figure 6 is an energy intensity result diagram of the dynamic response sensitive component of the structure to be measured provided by the embodiment of the application.

[0042] In the figure: 1-acceleration sensor, 2-core processor, 3-signal acquisition unit, 4-data preprocessing unit, 5-database unit, 6-response signal decomposition unit, 7-sensitive component selection unit, 8-sensitive component reconstruction unit, 9-damage index calculation unit, 10-early warning response unit, 11-remote communication early warning unit, 12-data storage playback unit, 13-data visualization unit, 14-user mobile terminal, 15-user computer terminal. DETAILED DESCRIPTION

[0043] In order to further illustrate the technical solutions disclosed in the application, the following will be described in detail in conjunction with the drawings and specific embodiments in the specification. Those skilled in the art should know that the preferred and improved made without departing from the spirit of the application fall within the protection scope of the application, and the conventional means and common techniques in the art are not described and explained in detail in this specific embodiment.

[0044] As shown in Figure 1 is a composition diagram of an engineering structure damage identification method and a lightweight monitoring system based on vibration response sensitive components; as Figure 2 is a specific implementation flow chart of an engineering structure damage identification method and device based on vibration response sensitive components; as Figure 3 is a flow chart of the engineering structure damage identification method based on the dynamic response sensitive component.

[0045] The following describes the technical solutions of the application in one specific embodiment:

[0046] A 12-story reinforced concrete frame structure is taken as a research object for analysis. The three-dimensional diagram and the plan view of the frame structure are as shown in Figure 4As shown, the structure is composed of columns, beams and floor slabs, the column height of the first floor structure column is 6.0 m, the cross-sectional size is 0.5 m x 0.5 m, the height of the remaining 11 layers of structure columns is 3.0 m, the cross-sectional size is 0.45 m x 0.45 m, the beam cross-sectional size is 0.3 m x 0.2 m, and the floor slab thickness is 0.1 m. The elastic modulus E of the material of all components is 32.5 GPa, the mass density p is 2500 kg / m3, and the Poisson's ratio μ is 0.2. The structure is mainly subjected to environmental random excitation, and the number of columns that may be damaged is as shown in Figure 4 As shown.

[0047] In this embodiment, the engineering structure damage identification method based on vibration response sensitive components and the lightweight monitoring system provided by the application are provided, which specifically includes the following steps:

[0048] Step 1: For the actual building structure monitoring object, a structure vibration sensing module is established for measuring the acceleration response of the structure facility monitoring position; a MEMS accelerometer is used as the main acceleration sensor (1) and is arranged in the middle position of the load-bearing column between adjacent two floors.

[0049] Specifically, MEMS accelerometer measuring points are arranged on all middle columns on the west outer side of the structure, as shown in Figure 4 A MEMS accelerometer of ADXL375 type is used to build the structure vibration sensing module.

[0050] Step 2: Design and develop a high-performance micro control module, which mainly consists of a core processor (2) and an embedded signal acquisition unit (3), a data preprocessing unit (4) and a database unit (5), for acquiring structure vibration response and transmitting to other core modules.

[0051] Specifically, a Raspberry Pi is used as the core processor (2) to build the high-performance micro control module.

[0052] Specifically, the signal acquisition unit (3) is connected with the structure vibration sensing module, ZigBee technology is used to realize wireless network transmission, and vibration response data acquisition function is realized.

[0053] Specifically, according to the monitoring structure dynamic characteristics, the sampling frequency and single sampling duration of the MEMS accelerometer are determined, the sampling frequency is set to 100 Hz, the single sampling duration is 1200 s, and the interval between two response sampling processes is set to 10 s.

[0054] Specifically, the data preprocessing unit (4) analyzes and processes the signal acquisition unit (3) output signal to obtain smooth and high-quality structural vibration response data, which is transmitted to the database unit (5) and the damage identification algorithm module respectively; wherein, the signal acquisition unit (3) output signal is mainly processed by zero mean value processing and trend item processing, the zero mean value processing refers to calculating the average value of single sampling response data, and then subtracting the average value from the single sampling response data, the trend item processing is to fit the linear trend item of single sampling response data by using the least square method and eliminate it;

[0055] Specifically, the database of the database unit (5) is established, which is used for temporarily storing the structural vibration response data;

[0056] Step 3: Deploy the damage identification algorithm module on the high-performance micro control module, which is composed of a response signal decomposition unit (6), a sensitive component selection unit (7), a sensitive component reconstruction unit (8) and a damage index calculation unit (9), which is used for analyzing and processing the low-level original response signal, obtaining the high-level feature data which can evaluate the structural damage condition, and judging the structural damage.

[0057] Specifically, taking the 5% stiffness reduction damage of No. 6 column as an example, the damage identification analysis is carried out.

[0058] Specifically, the engineering structure damage identification method based on vibration response sensitive component proposed in the application is packaged as a damage identification algorithm module program and deployed on the core processor (2), and the vibration response output by the data preprocessing unit (4) is called and analyzed by the damage identification algorithm module. The specific sub-steps of the structural damage identification method are as follows:

[0059] Step 3-1: According to the structure vibration sensing module in step 1, the MEMS accelerometer is arranged, the signal acquisition unit (3) collects the structural vibration response under the healthy state as the reference data, and the vibration response data of each measuring point is collected as the measured data in real time according to the sampling period set in step 2;

[0060] Step 3-2: Select the discrete Meyer function as the wavelet base function, determine the decomposition layer number as 6, and perform wavelet packet decomposition on the reference data and the measured data according to formula (1), and then reconstruct the last layer component according to formula (2); the vibration response data of all measuring points is decomposed and reconstructed into 64 wavelet packet components, denoted as According to formula (3), the energy values of all wavelet packet components are calculated, and according to formula (4), the energy matrices E of the wavelet packet components of the reference data and the measured data are constructed respectively j,h and E j,d ;

[0061]

[0062] wherein, are the wavelet packet coefficients of the adjacent layer nodes in response to the wavelet packet decomposition, h and g are low-pass and high-pass filters respectively, subscript i represents the wavelet packet component number, subscript k is the wavelet packet coefficient number, and l is the number of data points of the processed sequence.

[0063]

[0064] wherein, E n i represents the energy value of the i-th wavelet packet component of the n-th monitoring signal of the structure, and similarly, the energy matrix E j is named according to this rule.

[0065] Step 3-3: Extract and construct the energy vectors E j,h and E j,d of the same component of all monitoring points from the energy matrix E h and E d , and calculate the energy intensity vector D of any wavelet packet component according to formula (5), and then calculate the energy intensity information entropy value H wpt of any wavelet packet component according to formulae (6) and (7). Figure 5 The wavelet packet component numbers selected according to the entropy values in order are: 43, 62, 44, 41, 63, 40, 48, 61, 64, 49, 59, 46, 45 and 60, accounting for about 22% of the total frequency band;

[0066] D = log 10 (E d / E h ) (5)

[0067]

[0068] wherein, P b,c represents the proportion of the energy intensity value of the b-th wavelet packet component of the monitoring signal of the c-th monitoring point of the structure to the total sum of the energy intensity values of the b-th wavelet packet component of the monitoring signals of all monitoring points of the structure, D c b and D q b respectively represent the energy intensity values of the b-th wavelet packet component of the monitoring signals of the c-th monitoring point and the q-th monitoring point of the structure.

[0069] Step 3-4: According to formula (2), the vibration response sensitive component selected in step 3-3 is reconstructed by wavelet packet, the non-sensitive component information is removed, and the reconstructed signal data a h (t) and a d (t) of the reference data and the to-be-measured data are obtained.

[0070] Step 3-5: Calculate the sensitive component energy value E of the reconstruction signal data of each measuring point in the healthy state and the to-be-tested state according to formula (8) r,h and E r,d Calculate the energy intensity of the sensitive component of the to-be-tested structure vibration response as the damage index according to formula (9), as shown in Figure 6 The results show that the curve has a significant peak at the position of column No. 6, indicating that damage occurs at the position of column No. 6.

[0071]

[0072] S E = log 10 (E r,d / E r,h ) (9)

[0073] In the formula, E r represents the energy value of the reconstruction signal of the sensitive component of the monitoring signal of each measuring point, and S E represents the energy intensity value of the sensitive component of the monitoring signal of each measuring point of the to-be-tested structure.

[0074] Step 4: Create a cloud storage module, including a data storage playback unit (12) and a data visualization unit (13), for long-term storage of structure vibration response data and all high-level feature data of key time nodes and abnormal time nodes, and can be called and visually displayed; the monitoring data output by the data preprocessing unit (4) in step 2 and the feature results output by the damage identification algorithm module in step 3 are uploaded to the cloud storage module through the application programming interface (API) mode.

[0075] Step 5: Design and expand the structure damage warning module, which is composed of a warning response unit (10) and a remote communication warning unit (11), receives the high-level feature data of the damage identification algorithm module and determines whether to perform damage warning, and when the structure is identified to be damaged, the warning response unit (10) sends an on-site warning, and the remote communication warning unit (11) sends the warning information to the user end.

[0076] Specifically, the warning response unit (10) mainly includes two kinds of hardware, warning light and buzzer, which are externally connected with the core processor (2) and receive the start instruction of the data processing unit to issue an alarm including light and whistle on site.

[0077] Specifically, the remote communication warning unit (11) mainly includes a 5G communication module, which is externally connected with the core processor (2) and receives the start instruction of the damage identification algorithm module to send warning information to the user mobile device end.

[0078] Step 6: build user end for receiving early warning information and query, access structure historical monitoring data, divided into user mobile terminal (14) and user computer terminal (15). The data stored in the cloud storage server can be managed through API, webpage and other ways, such as online browsing, uploading, downloading, deleting, renaming, sharing and other operations.

[0079] The above embodiments are only preferred embodiments of the present application, and the protection scope of the present application is not limited to the above examples. Improvements and changes obtained by those skilled in the art without departing from the technical concept of the present application should also be considered as the protection scope of the present application.

Claims

1. A lightweight monitoring system for damage identification of engineering structures based on dynamic response sensitive components, characterized by: The steps for building the lightweight monitoring system include: Step 1: For the actual structural monitoring object, determine the key monitoring locations and key measurement points, and establish a structural vibration perception module; Step 2: Design and develop a high-performance micro-control module, establish a signal acquisition unit (3), a data pre-processing unit (4), and a database unit (5) according to functional requirements; connect the signal acquisition unit (3) with the structural vibration sensing module, and build a data transmission path among the signal acquisition unit (3), the data pre-processing unit (4), and the database unit (5); Step 3: Encapsulate the structural damage identification method based on the dynamic response sensitive component into a program, deploy it on a high-performance micro-control module, and establish a data transmission path between the data preprocessing unit (4) and the database unit (5); Step 4: Create a cloud storage module to enable the database unit (5) and the damage identification algorithm module to upload data to it, and store the dynamic response data of key time nodes and abnormal time nodes and the damage index result data; Step 5: Design and develop a structural damage warning module; Step 6: Build a client to receive warning information and query access to historical monitoring data; The specific steps of step 3 are as follows: Step 3-1: Based on the characteristics of the engineering structure and the analysis requirements, an acceleration sensor (1) is arranged near the monitoring location, and a signal acquisition unit (3) collects the structural dynamic response in a healthy state as the baseline data, and collects the dynamic response data of each measuring point in real time according to the set sampling time as the test data; Step 3-2: Select the wavelet basis function and determine the decomposition level. The wavelet basis function can be discrete Meyer function or dbN function. The decomposition level j is set to 3 to 8 levels. The reference data and the data to be measured are decomposed by wavelet packets according to formula (1). Then, the last layer component is reconstructed according to formula (2). The vibration response data of all measuring points are decomposed and reconstructed into i wavelet packet components, which are recorded as The energy values ​​of all wavelet packet components are calculated according to formula (3), and the energy matrices E of the wavelet packet components of the reference data and the data to be tested are constructed according to formula (4). j,h Hehe E j,d ; Where, are the wavelet packet coefficients of the adjacent layer nodes after the response is decomposed by the wavelet packet, h and g are low-pass and high-pass filters respectively, the subscript i represents the wavelet packet component number, the subscript k represents the wavelet packet coefficient number, and l is the number of processed sequence data points. Where, E n i Represents the energy value of the i-th wavelet packet component of the monitoring signal of the n-th measuring point on the structure. Similarly, the energy matrix E j The other elements in are named according to this rule; Step 3-3: Energy matrix E from wavelet packet components j,h and E j,d Extract and construct the energy vector E of the same component of all measurement points h and E d , and calculate the energy intensity vector of any wavelet packet component according to formula (5), and then calculate the energy intensity information entropy value H of any wavelet packet component according to formulas (6) and (7) wpt (b), finally, select the energy intensity information entropy value H in the wavelet packet component wpt (b) The smallest first 15% to 30% components are used as vibration response sensitive components; D. log 10 (E d / E h ) (5) Where, P b,c It represents the ratio of the energy intensity value of the bth wavelet packet component of the monitoring signal of the cth measuring point of the structure to the total energy intensity value of the bth wavelet packet component of the monitoring signal of all measuring points of the structure, D c b and D q b They represent the energy intensity values ​​of the bth wavelet packet component of the monitoring signal at the cth and qth measuring points of the structure respectively; Step 3-4: According to formula (2), the vibration response sensitive components selected in step 3-3 are reconstructed by wavelet packets, and the non-sensitive component information is eliminated to obtain the reconstructed signal data a of the reference data and the data to be measured. h (t) and a d (t); Step 3-5: Calculate the sensitive component energy value E of the reconstructed signal data of each measuring point in the healthy state and the test state according to formula (8): r,h and E r,d , according to formula (9), the energy intensity of the sensitive component of the vibration response of the structure to be tested is calculated as the damage index, and the damage condition of the structure to be tested is directly analyzed; S E =log 10 (AND r,d / AND r,h ) (9) Where, E r It represents the energy value of the signal reconstructed by the sensitive component of the monitoring signal at each measuring point, S E It represents the energy intensity value of the sensitive component in the monitoring signal of each measuring point of the structure to be measured.

2. The lightweight monitoring system for damage identification of engineering structures based on dynamic response sensitive components according to claim 1 is characterized by: It includes a structural vibration perception module, a high-performance micro-control module, a damage identification algorithm module, a cloud storage module, a structural damage warning module and a user-end module; the high-performance micro-control module is connected to the user-end module via the cloud storage module; the high-performance micro-control module is also connected to the structural damage warning module via the damage identification algorithm module, and the structural damage warning module is connected to the cloud storage module, and the structural vibration perception module serves as an input module of the high-performance micro-control module; The structural vibration sensing module uses an acceleration sensor as a sensing element to measure the acceleration response or displacement response of key monitoring positions and key measuring points of structural facilities; The high-performance micro-control module includes a core processor (2) and an embedded signal acquisition unit (3), a data pre-processing unit (4), and a database unit (5), and is used to collect structural dynamic responses and transmit them to other core modules, and to undertake edge computing functions; The damage identification algorithm module is deployed on a high-performance micro-control module and is composed of a response signal decomposition unit (6), a sensitive component selection unit (7), a sensitive component reconstruction unit (8) and a damage index calculation unit (9). It is used to analyze and process low-level original response signals, obtain high-level feature data that can evaluate the structural damage status, and make judgments on the structural damage. The cloud storage module includes a data storage and playback unit (12) and a data visualization unit (13), which are used to store the structural vibration response data and all high-level feature data of key time nodes and abnormal time nodes for a long time, and can be called and visualized; The structural damage warning module is composed of a warning response unit (10) and a remote communication warning unit (11), which receives high-level feature data from the damage identification algorithm module and determines whether to issue a damage warning. When damage to the structure is identified, the warning response unit (10) issues an on-site warning, and the remote communication warning unit (11) sends the warning information to the user end. User terminal module: divided into user mobile terminal (14) and user computer terminal (15), used to receive warning information and query and access structural historical monitoring data.

3. The lightweight monitoring system for damage identification of engineering structures based on dynamic response sensitive components according to claim 2 is characterized by: The structural vibration sensing module adopts a MEMS accelerometer.

4. The lightweight monitoring system for damage identification of engineering structures based on dynamic response sensitive components according to claim 2 is characterized by: The high-performance micro-control module is characterized in that the signal acquisition unit (3) is connected to the structural vibration sensing module to realize the dynamic response data acquisition function; the data preprocessing unit (4) analyzes and processes the output signal of the signal acquisition unit (3) to obtain stable, high-quality structural dynamic response data, and transmits the data to the database unit (5) and the damage identification algorithm module respectively; the database unit (5) is used to temporarily store the structural dynamic response data.

5. The lightweight monitoring system for damage identification of engineering structures based on dynamic response sensitive components according to claim 2 is characterized by: The signal acquisition unit (3) adopts ZigBee technology to realize wireless network transmission, and determines the sampling frequency and single sampling duration of the acceleration sensor (1) according to the dynamic characteristics of the monitoring structure. The sampling frequency is set to a frequency value greater than twice the fifth-order natural frequency value of the monitoring structure, the sampling duration is set to 600s to 3600s, and the interval between two response sampling processes is set to 5s to 10s.

6. The lightweight monitoring system for damage identification of engineering structures based on dynamic response sensitive components according to claim 2 is characterized by: The data preprocessing unit (4) performs zero mean processing and detrending term processing on the signal output by the signal acquisition unit (3); and the database unit (5) includes a database for storing short-term monitoring data.

7. The lightweight monitoring system for damage identification of engineering structures based on dynamic response sensitive components according to claim 1 is characterized by: The damage identification algorithm module is based on a structural damage identification method of a dynamic response sensitive component and is encapsulated as a program and deployed on a core processor (2), thereby realizing the function of performing real-time calculation and analysis of the structural dynamic response and judging the structural damage condition at the edge of a high-performance micro-control module.

8. The lightweight monitoring system for damage identification of engineering structures based on dynamic response sensitive components according to claim 2 is characterized by: The cloud storage module uploads data to the cloud storage module through the application programming interface API, web page or client; the user terminal module manages the data stored on the cloud storage server through API and web page.

9. The lightweight monitoring system for damage identification of engineering structures based on dynamic response sensitive components according to claim 2 is characterized by: In the structural damage warning module, the warning response unit (10) includes two hardware components: a high-brightness warning light and a high-volume horn, which are externally connected to the core processor (2) and receive the startup instruction of the damage identification algorithm module, and issue an alarm including lighting the light and sounding the horn on site; the remote communication warning unit (11) includes a 5G communication module, which is externally connected to the core processor (2) and receives the startup instruction of the damage identification algorithm module and issues a warning message to the user's mobile device.

Images