Intelligent monitoring method for state of self-adaptive supporting system of hanging basket of variable cross-section box girder

Abstract

The invention relates to the technical field of structural health monitoring, and discloses a variable cross-section box girder hanging basket adaptive support system state intelligent monitoring method, which comprises the following steps: synchronously acquiring structural response data and working condition fingerprint data of a hanging basket; constructing a health topology reference image set corresponding to each construction condition in an off-line manner, wherein the health topology reference image set is composed of a reference influence transfer coefficient matrix; during online monitoring, the current working condition is recognized in real time, an instantaneous influence transfer coefficient matrix is calculated, the difference between the instantaneous influence transfer coefficient matrix and a reference matrix of the corresponding working condition in a graph set is obtained, and a topological deformation matrix is generated. And further, the multi-dimensional state deviation of the system is quantified into a single risk index by calculating the structural topology entropy of the topology deformation matrix, and the single risk index is compared with a working condition self-adaptive threshold value. The technical problem that a traditional monitoring method is insensitive to systematic early degradation and cannot accurately position a fault source is solved, and high-sensitivity, quantitative evaluation and intelligent diagnosis of the overall health state of the hanging basket system are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of structural health monitoring, in particular to a state intelligent monitoring method for a variable cross-section box girder hanging basket adaptive support system. BACKGROUND

[0002] The cantilever pouring construction of the variable cross-section box girder is one of the core processes in the construction of modern large bridges, and the main bearing and operation platform is the hanging basket system. The hanging basket, as a large and temporary mobile bearing structure, integrates main trusses, forms, walking, anchoring and support, and other complex systems. In the construction process, the hanging basket needs to bear its own gravity, the huge and constantly changing load of newly poured concrete, and complete multi-condition cyclic operation. The stress state of its structure is extremely complex and variable. Therefore, real-time and effective safety monitoring of the hanging basket system is of great importance.

[0003] Currently, the monitoring methods for the hanging basket system mainly rely on manual inspection and automatic monitoring based on sensors. However, the existing automatic monitoring technology has significant limitations. These technologies usually focus on independent measurement of local physical quantities at key measurement points of the structure and compare the measurement values with a pre-set and fixed safety threshold. The inherent defect of this monitoring logic is that it can only respond to local and explicit risks that have developed to a certain extent, and it is not sensitive to systematic and progressive structural degradation or early damage. The overall health status of the structure is a comprehensive reflection of the mechanical correlation and transmission characteristics of all internal components. A small bolt loosening or the initial emergence of material fatigue may not immediately cause any dramatic changes in single-point readings, but it has already changed the mechanical transmission path and response mode of the entire system. Traditional point monitoring methods often fail to capture early signs of abnormalities at the system level due to the lack of analysis of the correlation between different parts of the structure, thus missing the best early warning opportunity.

[0004] In addition, when the readings of a certain measurement point do trigger an alarm, the existing technology also has difficulty in providing effective diagnostic information. The alarm only indicates where the problem is, but cannot explain why it happens and where the root cause is. The troubleshooting and positioning still highly depend on the experience of on-site engineers, and there is a lack of objective and intelligent tracing means. More importantly, the normal working state of the hanging basket itself includes multiple different construction conditions such as walking, form adjustment, concrete pouring, and prestress tensioning. The normal benchmarks of structural responses under different conditions are completely different. The existing monitoring methods generally lack the ability to automatically identify the construction conditions and adaptively adjust the evaluation benchmarks. The use of static and single threshold system is prone to generate a large number of false alarms or missed alarms when the conditions are switched, greatly reducing the reliability and practical value of the monitoring system. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides a variable cross-section box girder trolley self-adaptive support system state intelligent monitoring method, which solves the technical problems that the traditional monitoring technology is not sensitive to the early progressive degradation of the overall structure due to excessive reliance on independent threshold judgment of local measuring points and is difficult to effectively trace and diagnose when an anomaly occurs, and fundamentally solves the problems of monitoring false positives and false negatives caused by the inability of the evaluation benchmark to adaptively adjust to complex and changing construction conditions.

[0006] To achieve the above object, the present application provides a variable cross-section box girder trolley self-adaptive support system state intelligent monitoring method, comprising the following steps: Step S1, obtaining structural response data of structural response sensors installed at multiple positions on the trolley.

[0007] Step S2, based on a plurality of pre-set construction conditions, offline constructing a health topology benchmark atlas. The health topology benchmark atlas includes a benchmark influence transfer coefficient matrix corresponding to each construction condition. Specifically, for each construction condition, structural response data for calibration is collected under the healthy running state of the trolley, and the benchmark influence transfer coefficient matrix corresponding to the construction condition is calculated according to the data.

[0008] The elements in the benchmark influence transfer coefficient matrix are calculated by the following formula: ; Wherein, is an element in the benchmark influence transfer coefficient matrix representing the influence relationship between the i-th structural response sensor and the j-th structural response sensor; is any construction condition in the plurality of construction conditions; is the health calibration time period under the construction condition is an operator for obtaining the mathematical expectation within the time period is the change amount of the structural response data of the i-th and j-th structural response sensors within the preset time window is the mean value of the structural response data of the i-th and j-th structural response sensors within the preset time window

[0009] ​​​​​​​​​​​​Step S3, online monitoring stage, real-time identification of the current construction condition of the hanging basket, which can be based on the working condition fingerprint data of the working condition fingerprint sensor installed on the hanging basket or its associated equipment.

[0010] Step S4, according to the real-time acquired structure response data, calculate the instantaneous influence transfer coefficient matrix.

[0011] Step S5, according to the current construction condition identified in step S3, call the reference influence transfer coefficient matrix corresponding to the current construction condition from the health topology reference graph set constructed in step S2.

[0012] Step S6, compare the instantaneous influence transfer coefficient matrix calculated in step S4 with the reference influence transfer coefficient matrix called in step S5 to generate a topology deformation matrix, in the embodiment, the topology deformation matrix is obtained by subtracting the called reference influence transfer coefficient matrix from the instantaneous influence transfer coefficient matrix.

[0013] Step S7: based on the topology deformation matrix, calculate the structure topology entropy, which is used to converge the multi-dimensional topology deformation information into a single scalar to quantify the degree of deviation of the hanging basket system from its healthy state under the current working condition.

[0014] The structure topology entropy can be obtained by calculating the Frobenius norm of the topology deformation matrix: ; Wherein, is the structure topology entropy at time ; is the topology deformation matrix at time ; is the element corresponding to the first and the first structure response sensor in the topology deformation matrix; Indicates summing all elements of the matrix.

[0015] Step S8, according to the structure topology entropy, state evaluation of the hanging basket, the evaluation method includes comparing the structure topology entropy with the threshold preset for the current construction condition, and analyzing the time variation trend of the structure topology entropy to output warning or alarm information.

[0016] Step S9, when the state evaluation determines that the state of the hanging basket is abnormal, by analyzing the topology deformation matrix, locate one or more elements that contribute most to the structure topology entropy to determine the mechanical transmission path where the abnormality occurs.

[0017] The second aspect of the present application provides a variable cross-section box girder hanging basket self-adaptive support system state intelligent monitoring system, which comprises: A data acquisition module is configured to acquire structural response data of a plurality of structural response sensors on the hanging basket; A reference construction module is configured to construct a health topology reference atlas based on a plurality of preset construction conditions; The health topology reference atlas comprises a reference influence transmission coefficient matrix corresponding to each of the construction conditions; A real-time analysis module is configured to identify a current construction condition of the hanging basket in real time, and calculate an instantaneous influence transmission coefficient matrix according to the structural response data; A deformation generation module is configured to call a corresponding reference influence transmission coefficient matrix from the health topology reference atlas according to the current construction condition, and compare the instantaneous influence transmission coefficient matrix with the reference influence transmission coefficient matrix to generate a topology deformation matrix; A risk assessment module is configured to calculate a structural topology entropy based on the topology deformation matrix, and perform state assessment on the hanging basket according to the structural topology entropy.

[0018] The present application provides a variable cross-section box girder hanging basket adaptive support system state intelligent monitoring method. It has the following beneficial effects: 1、The present application constructs a health topology reference atlas adaptive to construction conditions, and compares the reference matching the current construction condition during monitoring. The present application can effectively distinguish the structural response data changes caused by normal condition switching from the data changes caused by real structural abnormalities, thereby reducing the possibility of misjudgment in state assessment.

[0019] 2、The present application monitors the structural topology entropy representing the stability of the internal mechanical relationship of the system, rather than the absolute value of the isolated physical quantity. It can capture the early signs of the evolution of the system from a stable state to an unstable state before the structural physical quantity reaches the preset danger threshold, thereby providing an earlier risk warning window.

[0020] 3、When the system state is abnormal, the present application can directly locate the main mechanical transmission path causing the increase of the structural topology entropy by analyzing the topology deformation matrix, thereby providing a clear technical guide for determining the physical root cause of the abnormal state and improving the directivity of fault troubleshooting. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The present application is a method flowchart; Figure 2 The present application is a system architecture diagram; Figure 3 The present application is a health topology reference atlas construction method flowchart; Figure 4 The present application is a sensor network layout schematic diagram; Figure 5 The online monitoring and topology deformation analysis method flowchart of the present application; Figure 6 The state evaluation and diagnosis method flowchart of the present application.

[0022] Wherein, 10, data acquisition module; 20, reference construction module; 30, real-time analysis module; 40, deformation generation module; 50, risk assessment module. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0024] EMBODIMENT: Please refer to the drawings of the embodiments of the present application Figure 1 - the drawings of the present application Figure 6 The embodiment of the present application provides a variable cross-section box girder hanging basket adaptive support system state intelligent monitoring method, which comprises: Step S100, multi-source data acquisition and synchronization.

[0025] In step S100 of the present application, multi-source heterogeneous data of the hanging basket system is collected and synchronized. The purpose of this step is to provide the original data basis for the subsequent health topology reference atlas construction and online state evaluation.

[0026] This step first lays out a sensor network, which includes a structural response sensor group and a working condition fingerprint sensor group. The structural response sensor group is used to obtain structural response data representing the mechanical behavior of the hanging basket under load. In the embodiment, the sensor group can include: Strain sensors laid out on the upper and lower chord bars and key web bars of the hanging basket main truss, used to measure the stress or strain of the bar; Pressure sensors or load sensors laid out at the front support point and rear anchoring area of the hanging basket, used to measure the support reaction force and anchoring force; Displacement sensors laid out at the cantilever end of the hanging basket, used to measure its vertical deflection; And inclination sensors laid out on the main truss, used to measure the overall longitudinal and transverse inclination of the hanging basket.

[0027] The working condition fingerprint sensor group is used to obtain working condition fingerprint data that can uniquely identify the current construction working condition of the hanging basket. In the embodiment, the sensor group can include: The current sensor is installed in the power supply circuit of the driving motor of the hanging basket walking mechanism, and the current reading of the sensor is significantly different under the conditions of the hanging basket being stationary and moving forward. The vibration sensor is installed on the concrete conveying pump or the distributing machine, and the vibration signal spectrum of the sensor has specific characteristics under the concrete pouring working condition. And the travel switch installed on the walking track of the hanging basket is used to accurately record the start and end positions of the hanging basket moving forward.

[0028] After the sensor network is laid out, the data of all sensors are synchronously collected by the data acquisition unit, the data acquisition unit is connected to a unified time source, such as a network time protocol server or a global positioning system clock module, to apply high-precision time stamps to all collected data points, ensuring strict alignment of data from different sources in the time dimension, and the sampling frequency of all channels is set to a fixed value, such as 50Hz, to ensure that the dynamic change process of the structural response can be captured.

[0029] The collected raw data needs to go through a preprocessing procedure before being stored in the database, and the preprocessing procedure includes: First, the signal is digitally filtered, such as using a Butterworth low-pass filter to filter out high-frequency noise introduced by environmental vibration or electromagnetic interference on the construction site; secondly, data cleaning is performed to identify and process outliers or null values in the data sequence caused by transient sensor failure, in the embodiment, outliers can be removed by setting a threshold, and short null values can be filled by linear interpolation.

[0030] Step S200, health topology reference atlas construction adaptive to working conditions.

[0031] In step S200 of the present application, health topology reference atlas construction adaptive to working conditions is performed, which is executed offline, and the purpose is to provide a health state reference model that can be compared and accurately corresponds to each construction condition for state evaluation in the online monitoring stage.

[0032] This step first divides and defines the construction conditions, according to the process flow of the variable cross-section box girder hanging basket cantilever construction, the whole working period of the hanging basket is divided into a limited, discrete set of construction conditions, in the embodiment, the set can include: The unloaded stationary condition, the hanging basket moving forward condition, the formwork installation and adjustment condition, the steel bar binding condition, the concrete pouring condition and the prestressed tensioning condition, the basis of this division is that each condition corresponds to a relatively stable or repetitive external load mode, so that the hanging basket structure produces a unique and identifiable mechanical response mode under this condition.

[0033] For each defined construction condition, corresponding health baseline data needs to be collected. This data collection process should be carried out when the hanging basket is first put into use or during the construction phase after safety verification, to ensure that the hanging basket structure is in a healthy state during the collection period. For each condition, a statistically significant structural response data for a sufficiently long period of time should be collected to form a health dataset for that condition.

[0034] Based on the collected health dataset, the baseline influence transmission coefficient matrix for each construction condition is calculated. (Baseline influence transmission coefficient) Its physical significance lies in the quantitative characterization of the construction conditions. Below, the first of the hanging basket structure The normalized perturbation at the location of the first sensor, for the first... The degree of influence of the normalized response generated at the location of each sensor is determined by the following formula: ; In this formula, It is the unique identifier for the current construction condition being calibrated. Under construction conditions The complete time period for collecting health baseline data.

[0035] and They are the first The and the first A structural response sensor at time The tiny time window centered on The change in reading within, in the embodiment, It is possible Calculated.

[0036] and They are the first The and the first A structural response sensor during this time window The purpose of introducing the average reading within the range for normalization is to eliminate the differences in physical dimensions and numerical ranges between different types of sensors, so that the calculation results only reflect the dimensionless proportional relationship between the rates of change of two physical quantities.

[0037] Represents the entire health calibration period Within the brackets, perform a mathematical expectation operation on the ratio. In practical calculations, this operation can be performed by determining the time period. It is achieved by the arithmetic mean of the ratios calculated at all discrete time points within the time frame. This time averaging operation can smooth out random noise in the data and extract stable values ​​that characterize the inherent mechanical transmission properties of the structure under this working condition.

[0038] all sensor pairs with physical causal correlation The above calculation is repeated to finally generate a complete benchmark influence transfer coefficient matrix corresponding to the working condition . .

[0039] Finally, all construction working conditions and their corresponding benchmark influence transfer coefficient matrices are stored in a structured manner to form a health topology benchmark atlas, which can be organized as a database or a hash table structure in the embodiment, wherein the unique identifier of each construction working condition is taken as the key and its corresponding benchmark influence transfer coefficient matrix is taken as the value. This storage structure ensures that in the subsequent online monitoring stage, the matrix can be quickly and accurately retrieved and called according to the working condition identifier identified in real time.

[0040] Step S300, online monitoring and topology deformation analysis.

[0041] In step S300 of the present application, online monitoring and topology deformation analysis are performed, which is executed in real time during the actual construction of the hanging basket, and the purpose is to obtain a topology deformation matrix representing the deviation of the current health state of the hanging basket.

[0042] This step first performs real-time working condition identification. The system continuously receives working condition fingerprint data from the working condition fingerprint sensor group, and based on the preset pattern recognition rule, at each time outputs the current construction working condition identifier In the embodiment, the identification rule can be based on the comparison of the reading value of one or more sensors with the preset threshold value, for example, the reading value of the current sensor installed on the driving motor of the hanging basket running mechanism is compared with the moving threshold value, when the reading value is greater than the threshold value, it is determined that the current working condition is the hanging basket forward moving working condition; when the reading value is less than the static threshold value, it is determined that the working condition is empty static working condition.

[0043] In another embodiment, a state machine model can be used for working condition identification, which predefines the logical conversion relationship between each construction working condition, for example, the system can only be converted to the template installation and adjustment working condition after the hanging basket forward moving working condition ends. This way takes advantage of the inherent timing of the construction process to improve the accuracy of working condition identification.

[0044] At the same time of determining the current working condition, the system calculates the instantaneous influence transfer coefficient matrix based on the real-time acquired structural response data, the formula used in the calculation is the same in form as the formula used in step S200 to calculate the benchmark coefficient, but there is an essential difference in the data window based on which the calculation is performed, here the calculation is performed in a short-time sliding window with a time length of , and the window changes with time It slides forward as time goes on. This calculation does not perform long-term statistical averaging, and its results... This reflects the constant state of the hanging basket structure. The dynamic correlation characteristics of the vicinity within this short period of time.

[0045] Subsequently, the system generates the topology deformation matrix based on the current construction condition identifier output by the real-time condition identification module. The system uses this identifier as an index to retrieve and call up the corresponding baseline influence transmission coefficient matrix from the health topology baseline map set constructed in step S200. .

[0046] Obtain the instantaneous matrix and benchmark matrix Then, by performing element-wise subtraction on the two matrices, a topological deformation matrix is ​​generated. : ; Each element of the topological deformation matrix Each has a clear physical meaning; its magnitude and sign directly quantify the changes from the first... The location of the sensor is from the first... The deviation of the mechanical transmission path at the location of each sensor from its healthy state under the same construction conditions at the current moment is represented by the following values: a value close to zero indicates that the path is performing normally; a significant positive value indicates that the mechanical influence transmission of the path is abnormally enhanced; and a significant negative value indicates that the mechanical influence transmission of the path is abnormally weakened.

[0047] Step S400: Quantification of structural topological entropy and state assessment and diagnosis.

[0048] In step S400 of the present invention, structural topological entropy quantification and state assessment diagnosis are performed. This step receives the topological deformation matrix generated in step S300 as input, aiming to converge multi-dimensional deviation information into a single quantitative risk indicator, and perform state assessment and anomaly tracing accordingly.

[0049] This step first calculates the structural topological entropy, a scalar used to measure the degree to which the overall topological deformation matrix deviates from the zero matrix. In this embodiment, the structural topological entropy... By applying the topological deformation matrix To achieve this, calculate its Frobenius norm: ; in, For at any time The structural topological entropy; For at any time topology deformation matrix; corresponding to the first and the second structural response sensors; represents the sum of squares of all elements in the matrix, which squares and accumulates the deviation of all mechanical transmission paths, and the result can sensitively reflect any slight change in the overall state of the system.

[0050] After obtaining the time series of structural topology entropy, state evaluation is performed, which can include threshold-based judgment and trend-based analysis. For threshold-based judgment, the system presets attention threshold and alarm threshold for each construction condition . These thresholds can be obtained by statistical analysis of the structural topology entropy sequence calculated from the health data set of each condition in step S200. For example, the mean value of the health topology entropy under a certain condition plus twice the standard deviation can be used as the attention threshold, and the mean value plus three times the standard deviation can be used as the alarm threshold. When the attention threshold is exceeded, the system outputs a warning message; when the alarm threshold is exceeded, the system outputs an alarm message.

[0051] For trend-based analysis, the system calculates the first derivative or slope of the structural topology entropy time series within a sliding time window. When the slope remains positive and exceeds the preset trend threshold for a certain period of time, even if the absolute value has not reached the attention threshold, the system can output a warning message. This analysis is used to identify the gradual and slow deterioration process of the structure state.

[0052] When the state evaluation module outputs an alarm message, the system automatically triggers a diagnostic tracing process, which retrieves the topology deformation matrix at the time of triggering the alarm. The system traverses all elements of the matrix and sorts them according to their absolute values, outputting one or more elements with the largest absolute values.

[0053] Each of the outputted elements with the largest absolute values has a corresponding sensor index pair , which indicates the mechanical transmission path that has the most significant contribution to the overall state of the system. This path is a mechanical influence transmission channel from the physical location of the first sensor to the physical location of the second sensor. This information can be directly used to guide the on-site engineering personnel to conduct targeted investigation on the specific physical part of the hanging basket.

[0054] Refer to the attached Figure 2, the variable cross-section box girder trolley self-adaptive support system state intelligent monitoring system, the system comprises a data acquisition module 10, a benchmark construction module 20, a real-time analysis module 30, a deformation generation module 40 and a risk assessment module 50.

[0055] The data acquisition module 10 is used for performing the function of step S100, the module comprises a data interface physically connected with the structural response sensor group and the working condition fingerprint sensor group, and a data acquisition unit with a built-in high-precision clock, the data acquisition unit is responsible for analog-to-digital conversion and synchronous sampling of all sensor channels at a preset sampling frequency, and a unified time stamp is attached to each data point, the data acquisition module 10 further comprises a data preprocessing unit, the unit is configured with a digital filter and a data cleaning algorithm, and is used for filtering and denoising and abnormal value processing on the collected original data.

[0056] The benchmark construction module 20 is used for performing the function of step S200, the module receives each working condition calibration data set collected and preprocessed under the trolley health state from the data acquisition module 10, and the module comprises a calculation unit inside, the calculation unit is configured to calculate the benchmark influence transfer coefficient for each construction working condition data set, the calculation unit can perform change amount, mean value, normalized ratio and mathematical expectation operation on the data of any specified sensor pair within a specified time period, the benchmark construction module 20 further comprises a storage management unit, which is used for associating each benchmark influence transfer coefficient matrix generated by calculation with the corresponding construction working condition identifier, and solidifying storage to form a health topology benchmark graph set.

[0057] The real-time analysis module 30 is used for performing part of the function of the online monitoring stage, the module comprises a working condition identification unit and an instantaneous matrix calculation unit, the working condition identification unit continuously receives real-time working condition fingerprint data from the data acquisition module 10, and executes preset pattern recognition rules, such as threshold logic or state machine model, to output the current construction working condition identifier at each time, and the instantaneous matrix calculation unit continuously receives real-time structural response data from the data acquisition module 10, and executes the same core calculation as the benchmark construction module 20 within a short-time sliding time window to generate an instantaneous influence transfer coefficient matrix representing the current dynamic characteristics of the trolley.

[0058] The deformation generation module 40 is used to perform another part of the function of the online monitoring stage, the input end of the module is connected to the real-time analysis module 30 to receive the current construction condition identifier and the instantaneous influence transfer coefficient matrix output by the real-time analysis module 30, the deformation generation module 40 internally includes a retrieval unit and a matrix operation unit, the retrieval unit searches in the health topology benchmark set generated by the benchmark construction module 20 according to the received condition identifier, and calls the corresponding benchmark influence transfer coefficient matrix, and the matrix operation unit performs element-by-element subtraction operation on the instantaneous influence transfer coefficient matrix and the called benchmark influence transfer coefficient matrix, and the operation result is the topology deformation matrix.

[0059] The risk assessment module 50 is used to perform the final state evaluation and diagnosis on the hanging basket, the module receives the topology deformation matrix from the deformation generation module 40, the module internally includes an entropy calculation unit, a state evaluation unit and a diagnosis tracing unit, the entropy calculation unit performs the Frobenius norm operation on the input topology deformation matrix, and outputs the structure topology entropy in the form of a single scalar, the state evaluation unit compares the structure topology entropy with the preset threshold value stored in the internal unit for the current condition, and combines the trend analysis on the entropy value time sequence to output the current state level of the hanging basket, such as normal, pre-warning or alarm, when the state is alarm, the diagnosis tracing unit is activated, the unit analyzes the topology deformation matrix triggering the alarm, searches and locates the element with the maximum absolute value in the matrix, and outputs the sensor index pair corresponding to the element to indicate the abnormal mechanical transmission path.

Claims

1. A method for intelligent monitoring of the state of a self-adaptive support system for a variable cross-section box girder trolley, characterized in that, The method comprises the following steps: obtaining structural response data of a plurality of structural response sensors on the hanging basket; constructing a health topology benchmark set based on a plurality of preset construction conditions; the health topology benchmark set comprises a benchmark influence transfer coefficient matrix corresponding to each construction condition; real-time identification of the current construction condition of the hanging basket; calculating an instantaneous influence transfer coefficient matrix according to the structural response data; calling the corresponding benchmark influence transfer coefficient matrix from the health topology benchmark set according to the current construction condition; comparing the instantaneous influence transfer coefficient matrix with the benchmark influence transfer coefficient matrix to generate a topology deformation matrix; calculating a structural topology entropy based on the topology deformation matrix; state evaluation of the hanging basket according to the structural topology entropy.

2. The method of claim 1, wherein the method further comprises: The step of constructing the health topology benchmark set comprises: for each construction condition, collecting calibration structural response data under the health operating state of the hanging basket; calculating the benchmark influence transfer coefficient matrix corresponding to the construction condition based on the calibration structural response data.

3. The method of claim 2, wherein the method further comprises: The elements in the benchmark influence transfer coefficient matrix are calculated by the following formula: ； wherein, is an element in the influence transmission coefficient matrix representing the influence relationship between the jth structural response sensor and the kth structural response sensor; is an element in the influence transmission coefficient matrix representing the influence relationship between the jth structural response sensor and the kth structural response sensor; is an element in the influence transmission coefficient matrix representing the influence relationship between the jth structural response sensor and the kth structural response sensor; is any construction condition in the plurality of construction conditions; is a health calibration time period under the construction condition is a health calibration time period under the construction condition is a health calibration time period under the construction condition is an operator for obtaining a mathematical expectation within the time period is an operator for obtaining a mathematical expectation within the time period is an operator for obtaining a mathematical expectation within the time period is an operator for obtaining a mathematical expectation within the time period is an operator for obtaining a mathematical expectation within the time period is a variation of the structural response data of the jth structural response sensor and the kth structural response sensor within a preset time window is a variation of the structural response data of the jth structural response sensor and the kth structural response sensor within a preset time window is a variation of the structural response data of the jth structural response sensor and the kth structural response sensor within a preset time window is a variation of the structural response data of the jth structural response sensor and the kth structural response sensor within a preset time window is a variation of the structural response data of the jth structural response sensor and the kth structural response sensor within a preset time window is a variation of the structural response data of the jth structural response sensor and the kth structural response sensor within a preset time window 4. The method of claim 1, wherein the method further comprises: The step of real-time identification of the current construction condition of the hanging basket comprises: obtaining working condition fingerprint data of one or more working condition fingerprint sensors installed on the hanging basket or its associated equipment; determining the current construction condition by using a preset pattern recognition rule based on the working condition fingerprint data.

5. The method of claim 1, wherein, The elements in the instantaneous influence transfer coefficient matrix and the benchmark influence transfer coefficient matrix both represent the ratio relationship between the normalized change rate of one structural response sensor and the normalized change rate of another structural response sensor.

6. The method of claim 1, wherein the method further comprises: The topology deformation matrix is obtained by subtracting the called benchmark influence transfer coefficient matrix from the instantaneous influence transfer coefficient matrix.

7. The method of claim 1, wherein the method further comprises: The structural topology entropy is obtained by calculating the Frobenius norm of the topology deformation matrix. ； wherein, is the structural topological entropy at time ; is the topological deformation matrix at time ; are the elements of the topological deformation matrix corresponding to the first and the first structural response sensors; denotes the sum over all elements of a matrix.

8. The method of claim 1, wherein the method further comprises: The step of state evaluation of the hanging basket according to the structural topology entropy comprises: comparing the structural topology entropy with a preset threshold value for the current construction condition or analyzing the time variation trend of the structural topology entropy to output early warning or alarm information.

9. The method of claim 1, wherein the method further comprises: The method further comprises: when the state evaluation determines that the state of the hanging basket is abnormal, locating one or more elements that contribute most to the structural topology entropy by analyzing the topology deformation matrix to determine the mechanical transmission path where the abnormality occurs.

10. The intelligent monitoring system for the state of the self-adaptive support system of the variable cross-section box girder trolley according to any one of claims 1-9, characterized in that, comprise: a data acquisition module for obtaining structural response data of a plurality of structural response sensors on the hanging basket; a benchmark construction module for constructing a health topology benchmark set based on a plurality of preset construction conditions; the health topology benchmark set comprises a benchmark influence transfer coefficient matrix corresponding to each construction condition; a real-time analysis module for real-time identification of the current construction condition of the hanging basket and calculation of an instantaneous influence transfer coefficient matrix according to the structural response data; a deformation generation module configured to call a corresponding reference influence transmission coefficient matrix from the health topology reference set according to the current construction condition, and compare the instantaneous influence transmission coefficient matrix with the reference influence transmission coefficient matrix to generate a topology deformation matrix; a risk assessment module configured to calculate a structure topology entropy based on the topology deformation matrix, and perform state assessment on the hanging basket according to the structure topology entropy.

Images