Structural Reliability Analysis Method and System Based on Multidimensional Data Analysis

Abstract

The present invention discloses a structural reliability analysis method and system based on multi-dimensional data analysis, which relates to the technical field of structural reliability analysis. A multi-dimensional data reference template is constructed, and in-depth analysis is carried out on historical structural reliability data to extract historical structural parameter groups and their corresponding state change data. According to the structural state change data, the historical structural parameter groups are classified to form a historical structural parameter set, and further divided into historical structural parameter group clusters. The structural parameters are obtained in real time to form a real-time structural parameter group, and compared and matched with the historical parameter groups according to the priority reference order. By selecting the most matching historical parameter group and using the associated structural state change data, the reliability of the current entity structure is accurately evaluated. The above technical solution of the present invention realizes the reliability evaluation of the entity structure and provides data support for ensuring the safe operation of the entity structure.

Description

Technical Field

[0001] The present invention relates to the technical field of structural reliability analysis, and in particular to a structural reliability analysis method and system based on multi-dimensional data analysis. Background Art

[0002] In the design and operation and maintenance process of modern engineering structures, ensuring the safety, stability and durability of the structure is of crucial importance. And the accurate acquisition and effective analysis of structural parameters are the core links for evaluating structural reliability. Traditionally, structural reliability assessment mainly relies on limited and single-dimensional structural parameter data, such as stress, displacement, etc. These methods often have difficulty in comprehensively capturing the complex behavior characteristics of the structure under different working conditions. With the continuous progress of sensor technology, Internet of Things technology and data processing technology, it has become possible to obtain multi-dimensional parameter data of the structure (including but not limited to multi-dimensional information such as stress, displacement, acceleration, temperature, humidity, vibration frequency, etc.).

[0003] However, how to efficiently integrate, process and analyze this massive, heterogeneous and high-dimensional structural parameter data to achieve accurate assessment of structural reliability has become a major challenge faced by the current engineering field. Summary of the Invention

[0004] The object of the present invention is to provide an analysis method and system capable of evaluating structural reliability.

[0005] The present invention discloses a structural reliability analysis method based on multi-dimensional data analysis, including:

[0006] Constructing a multi-dimensional data reference template, and based on the multi-dimensional data reference template, analyzing historical structural reliability record data to extract several historical structural parameter groups and the structural state change data corresponding to each historical structural parameter group;

[0007] Based on the structural state change data, classifying the historical structural parameter groups to form several historical structural parameter sets, and randomly comparing each historical structural parameter set according to its respective structural state change data and associating them according to the degree of equivalence to form several historical structural parameter group clusters in the historical structural parameter sets;

[0008] Determining the priority reference order of each parameter group based on the number of parameter groups in the historical structural parameter group clusters;

[0009] Based on a multi-dimensional data reference template, real-time structure parameters are obtained in real-time to form a real-time structure parameter group. Based on the priority reference order of the historical structure parameter group, the real-time structure parameter group is compared and matched with the historical structure parameter group in sequence. Based on the matching result, the most consistent historical structure parameter group is selected, and based on the structure state change data associated with the most consistent historical structure parameter group, the reliability of the entity structure corresponding to the real-time structure parameter group is evaluated.

[0010] In some embodiments disclosed in the present invention, the method for constructing a multi-dimensional data reference template includes:

[0011] Perform three-dimensional structure analysis on the entity structure corresponding to the historical structure reliability record data, and respectively determine the connection block, turning block, main load-bearing block, and conventional structure block on the entity structure;

[0012] Construct a three-dimensional analysis space, set a positioning coordinate system for the three-dimensional analysis space, configure the entity structure in the historical structure reliability record data in the positioning coordinate system, and configure the determined connection block, turning block, main load-bearing block, and conventional structure block on the positioning coordinate system;

[0013] Record the relative position characteristics between the connection block, turning block, main load-bearing block, and conventional structure block, determine and record their respective historical structure parameters, set the relative position characteristics on the positioning coordinate system, and construct a multi-dimensional data reference template, where several historical structure parameters are set for each block, and all the historical structure parameters corresponding to all blocks are recorded as the historical structure parameter group.

[0014] In some embodiments disclosed in the present invention, the expression forms of the relative position characteristics between the connection block, turning block, main load-bearing block, and conventional structure block include:

[0015] Respectively determine the connection block, turning block, main load-bearing block, and conventional structure block, and randomly select several block random mapping points for each block respectively, and calculate the average value of the block random mapping points of each block, which is recorded as the block reference mapping point;

[0016] Connect the block reference mapping points corresponding to each block to form a block relative position expression model.

[0017] In some embodiments disclosed in the present invention, the method for analyzing the historical structure reliability record data based on the multi-dimensional data reference template includes:

[0018] Analyze the reliability record data of historical structures to determine the historical structure templates, historical structure parameter groups, and structure state change data corresponding to the historical single-record data. Based on the matching relationship between the multi-dimensional data reference template and the historical structure template, determine the called historical structure parameter group and structure state change data;

[0019] Among them, the methods for determining the matching relationship between the multi-dimensional reference template and the historical structure template include:

[0020] Compare the multi-dimensional reference template and the historical structure template, and determine whether the relative position characteristics between the corresponding connection blocks, turning blocks, main load-bearing blocks, and conventional structure blocks match. If they match, it is considered that the historical structure template matches the multi-dimensional reference template;

[0021] Among them, the methods for determining whether the relative position characteristics match include:

[0022] Compare the block relative position expression models of the multi-dimensional reference template and the historical structure template. For each block reference mapping point, randomly select several adjacent block reference mapping points, calculate the mapping point distance between the block reference mapping point and the adjacent block reference mapping points, and calculate the average value of the mapping point distances, denoted as the average mapping point distance, and associate the average mapping point distance with the block reference mapping point;

[0023] Determine the difference distance of each pair of corresponding block reference mapping points, and based on the average mapping point distance corresponding to the block reference mapping point, determine the sub-matching parameter corresponding to the block reference mapping point. If the sub-matching parameter is greater than or equal to the preset value, it is considered that the corresponding block reference mapping point matches, and based on the proportion of the matching block reference mapping points, determine the matching parameter of the relative position characteristics between the multi-dimensional reference template and the historical structure template. If the matching parameter is greater than or equal to the preset value, it is considered that the relative position characteristics match;

[0024] The expression for calculating the matching parameter of the relative position characteristics is: ;

[0025] Among them, W is the matching parameter, J is the matching parameter conversion coefficient, is the sub-matching parameter judgment function of the i-th pair of corresponding block mapping points. If the sub-matching parameter is greater than or equal to the preset value, then output 1, otherwise output 0, n is the total number of all pairs of corresponding block mapping points, is the number of block mapping points of the block relative position expression model corresponding to the multi-dimensional reference template, n is the number of block mapping points of the block relative position expression model corresponding to the historical structure template, c is the block mapping point matching influence adjustment constant, and R is the block mapping point matching influence adjustment coefficient;

[0026] Among them, the expression for calculating the sub-matching parameter is: ;

[0027] Among them, is the sub-matching parameter of the i-th pair of corresponding block reference mapping points, is the preset maximum sub-matching parameter, is the difference distance of the i-th pair of corresponding block reference mapping points, is the average distance between mapping points corresponding to the i-th pair of corresponding block reference mapping points, is the preset standard distance between mapping points, and f is the mapping point distance influence adjustment coefficient.

[0028] In some embodiments disclosed in the present invention, the method for classifying historical structure parameter groups based on structure state change data includes:

[0029] Analyze the structure state change data corresponding to each historical structure parameter, determine several structure factor state change data, and analyze each structure factor state change data to determine the structure factor state parameters at different time nodes, and construct a structure factor state parameter change curve. Denote the combination of structure factor state parameter change curves belonging to the same structure state change data as a structure factor state parameter change curve group;

[0030] Classify the structure factor state parameter change curve groups according to the similarity relationship between the structure factor state parameter change curve groups, and classify the historical structure parameter groups corresponding to the structure factor state parameter change curve groups according to the classification of the structure factor state parameter change curve groups;

[0031] Among them, the method for determining the similarity relationship between the structure factor state parameter change curve groups includes:

[0032] Align the corresponding structure factor state parameter change curves, and compare the curvature integral values of several preset equivalent time periods. If the difference amount of the curvature integral values is within the preset interval, it is considered that the corresponding structure factor state parameter change curves are similar to each other;

[0033] Determine the number of similar curves of the structure factor state parameter change curves with a mutual similarity relationship. If the ratio of the number of similar curves to the total number of all structure factor state parameter change curves is greater than or equal to the preset value, it is considered that there is a similarity relationship between the structure factor state change curve groups.

[0034] In some embodiments disclosed by the present invention, the method for determining the equivalence degree between structure state change data includes:

[0035] Compare the curve groups of structure factor state parameters corresponding to the structure state change data. For the curve of structure factor state parameters, a number of curve probe points are set. Calculate the curve probe point difference values corresponding to each curve probe point respectively. If the curve probe point difference value is less than or equal to the preset value, it is determined that the corresponding curve probe point is equivalent, denoted as an equivalent curve probe point;

[0036] If the ratio of the number of equivalent curve probe points belonging to the same curve of structure factor state parameters to the number of all curve probe points belongs to the first preset ratio interval, mark the first sub - equivalence degree for the corresponding curve of structure factor state parameters. If the ratio of the number of probe points belongs to the second preset ratio interval, mark the second sub - equivalence degree for the corresponding curve of structure factor state parameters. If the ratio of the number of probe points belongs to the third preset ratio interval, mark the third sub - equivalence degree for the corresponding curve of structure factor state parameters,..., if the ratio of the number of probe points belongs to the nth preset ratio interval, mark the nth sub - equivalence degree for the corresponding curve of structure factor state parameters;

[0037] Accumulate and sum up the sub - equivalence degrees corresponding to all the curves of structure factor state parameters to obtain the equivalence degree between the structure state change data.

[0038] In some embodiments disclosed by the present invention, the method for comparing and matching the real - time structure parameter group and the historical structure parameter group includes:

[0039] Set attention weight coefficients for different structure parameters, and combine the structure parameter differences of each structure parameter between the structure parameter groups to determine the sub - structure difference parameters corresponding to the structure parameters. Based on the sub - structure difference parameters corresponding to all the structure parameters, determine the structure difference parameter between the real - time structure parameter group and the historical structure parameter group;

[0040] If the structure difference parameter is less than or equal to the preset value, it is determined that the historical structure parameter group is the most matching historical structure parameter group;

[0041] Among them, the expression for calculating the structure difference parameter is: ;

[0042] Among them, Y is the structure difference parameter, is the sub - structure difference parameter of the xth structure parameter, is the attention weight coefficient of the xth structure parameter, and N is the number of structure parameters.

[0043] In some embodiments disclosed by the present invention, the method for evaluating the reliability of the entity structure corresponding to the real - time structure parameter group based on the structure state change data includes:

[0044] Analyze the data of the structural state change, determine several sets of data of the structural factor state change. For each set of data of the structural factor state change, several state change intervals are set, and corresponding reliability parameters are set for each state change interval. Based on the state change interval to which each set of data of the structural factor state change belongs, determine the reliability parameter corresponding to the set of data of the structural factor state change;

[0045] Based on the reliability parameters corresponding to different sets of data of the structural factor state change, evaluate the reliability of the entity structure.

[0046] In some embodiments disclosed by the present invention, the present invention also discloses a structural reliability analysis system based on multi-dimensional data analysis, including:

[0047] The first module constructs a multi-dimensional data reference template, and based on the multi-dimensional data reference template, analyzes the historical structural reliability record data, extracts several historical structural parameter groups, and the data of the structural state change corresponding to each historical structural parameter group;

[0048] The second module is used to classify the historical structural parameter groups based on the data of the structural state change, form several historical structural parameter sets, and randomly compare each historical structural parameter set according to its own data of the structural state change, and associate them according to the degree of equivalence, and form several clusters of historical structural parameter groups in the historical structural parameter sets;

[0049] The third module is used to determine the priority reference order of each parameter group based on the number of parameter groups in the clusters of historical structural parameter groups in the historical structural parameter group clusters;

[0050] The fourth module is used to obtain real-time structural parameters in real time based on the multi-dimensional data reference template, form real-time structural parameter groups, and based on the priority reference order of the historical structural parameter groups, compare and match the real-time structural parameter groups with the historical structural parameter groups in turn, and based on the matching results, select the most matching historical structural parameter group, and evaluate the reliability of the entity structure corresponding to the real-time structural parameter group based on the data of the structural state change associated with the most matching historical structural parameter group.

[0051] The present invention discloses a structural reliability analysis method and system based on multi-dimensional data analysis, which relates to the technical field of structural reliability analysis. A multi-dimensional data reference template is constructed, and in-depth analysis is carried out on historical structural reliability data to extract historical structure parameter groups and their corresponding state change data; according to the structural state change data, the historical structure parameter groups are classified to form a historical structure parameter set, and further divided into historical structure parameter group clusters; structural parameters are obtained in real time to form a real-time structure parameter group, and comparison and matching are carried out with the historical parameter groups according to the priority reference order; by selecting the most matching historical parameter group and using the associated structural state change data, the reliability of the current entity structure is accurately evaluated. The above technical solution of the present invention realizes the reliability evaluation of the entity structure and provides data support for ensuring the safe operation of the entity structure.

[0052] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a method step diagram of the structural reliability analysis method based on multi-dimensional data analysis disclosed in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0055] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention and should not be construed as limiting the protection scope of the present invention. Those skilled in the art can make some non-essential improvements and adjustments based on the content of the present invention described below. In the present invention, unless otherwise clearly defined and limited, the technical terms used in the present invention should have the ordinary meaning understood by those skilled in the art of the present invention.

[0056] Embodiment:

[0057] The present invention discloses a structural reliability analysis method based on multi-dimensional data analysis. Refer to Figure 1 , including:

[0058] Step S100, construct a multi-dimensional data reference template, and based on the multi-dimensional data reference template, analyze the historical structural reliability record data to extract several historical structure parameter groups and the structural state change data corresponding to each historical structure parameter group.

[0059] First, construct a multi-dimensional data reference template that only contains the basic geometric information of the structure. This template provides a clear geometric framework for the entity structure, serving as the basis for defining and standardizing subsequent data collection and analysis. Based on this template, conduct in-depth mining of historical structural data, with particular attention to the changes in structural data over time, i.e., the structural state change data. Such data includes, but is not limited to, information on the evolution of structural deformation, crack development, material aging, etc. over time. Through this step, the aim is to establish a database containing a rich historical record of structural state changes, providing a solid data foundation for subsequent analysis.

[0060] In some embodiments disclosed in the present invention, the method for constructing the multi-dimensional data reference template includes:

[0061] Step S101, perform three-dimensional structural analysis on the entity structure corresponding to the historical structural reliability record data, and respectively determine the connection block, turning block, main load-bearing block, and conventional structural block on the entity structure.

[0062] First, conduct in-depth analysis of the historical structural reliability record data, which is the basis for constructing the multi-dimensional data reference template. Through three-dimensional structural analysis techniques, key blocks on the entity structure can be identified and determined, including connection blocks (i.e., connection parts in the structure, such as welds, bolt connections, etc.), turning blocks (parts where the structural shape or direction changes significantly), main load-bearing blocks (parts that bear the main load or stress), and conventional structural blocks (other parts except the above special blocks). The determination of these blocks helps to more accurately understand the overall performance and potential weaknesses of the structure.

[0063] Step S102, construct a three-dimensional analysis space, set a positioning coordinate system for the three-dimensional analysis space, configure the entity structure in the historical structural reliability record data in the positioning coordinate system, and configure the determined connection block, turning block, main load-bearing block, and conventional structural block on the positioning coordinate system.

[0064] Next, construct a three-dimensional analysis space and set a positioning coordinate system for it. This coordinate system is the basis for subsequent data positioning and analysis. Configuring the entity structure in the historical structural reliability record data into this positioning coordinate system means corresponding the actual position and form of the structure with the coordinate system. At the same time, configuring each block determined in step S101 onto the positioning coordinate system allows the positions and relationships of each block to be visually seen in three-dimensional space.

[0065] Step S103: Record the relative position features among the connection block, the transition block, the main load-bearing block, and the conventional structure block, determine and record their respective historical structure parameters, set the relative position features on the positioning coordinate system, and construct a multi-dimensional data reference template. Each block is set with several historical structure parameters, and all the historical structure parameters corresponding to all blocks are recorded as the historical structure parameter group.

[0066] In this step, recording the relative position features among the blocks is crucial for understanding the overall layout of the structure and the interaction between the blocks. At the same time, it is also necessary to determine and record the historical structure parameters of each block, which may include material properties, dimensions, shapes, load histories, etc. By setting the relative position features and the historical structure parameters of the blocks on the positioning coordinate system, a multi-dimensional data reference template is constructed. This template not only contains the geometric information of the structure but also integrates the historical performance data of the structure, providing rich data support for subsequent structure analysis, evaluation, and optimization.

[0067] In addition, each block is set with several historical structure parameters, and these parameters together constitute the structural feature description of the block. All the historical structure parameters corresponding to all blocks are combined to form the historical structure parameter group. This parameter group is the basic data set for subsequent structure comparison, status evaluation, and other work.

[0068] In some embodiments disclosed by the present invention, the expression forms of the relative position features among the connection block, the transition block, the main load-bearing block, and the conventional structure block include:

[0069] Step S1031: Determine the connection block, the transition block, the main load-bearing block, and the conventional structure block respectively, randomly select several block random mapping points for each block respectively, and calculate the average value of the block random mapping points of each block, which is denoted as the block reference mapping point.

[0070] First, it is necessary to accurately determine the connection block, the transition block, the main load-bearing block, and the conventional structure block. This is the basis for constructing the relative position feature expression model. After determining these blocks, randomly select several block random mapping points for each block. These mapping points can be any points on the surface or inside the block and are used to represent the position of the block in space.

[0071] Next, calculate the average value of the block random mapping points of each block. This average value is called the block reference mapping point. The block reference mapping point can be regarded as a "center of gravity" or "representative point" of the block in three-dimensional space, and it can accurately reflect the position features of the block in space.

[0072] By selecting random mapping points and calculating the average value, the error caused by improper selection of a single mapping point can be effectively reduced, and the accuracy of the expression of the block position characteristics can be improved.

[0073] Step S1032: Connect the block reference mapping points corresponding to each block to form a block relative position expression model.

[0074] After obtaining the block reference mapping points of each block, these reference mapping points are connected to form a block relative position expression model. This model visually shows the relative position relationship between each block.

[0075] In some embodiments disclosed by the present invention, the method for analyzing historical structure reliability record data based on a multi-dimensional data reference template includes:

[0076] Step S104: Analyze the historical structure reliability record data to determine the historical structure template, historical structure parameter group, and structure state change data corresponding to the historical single record data, and determine the called historical structure parameter group and structure state change data based on the matching relationship between the multi-dimensional data reference template and the historical structure template.

[0077] Among them, the method for determining the matching relationship between the multi-dimensional reference template and the historical structure template includes:

[0078] Compare the multi-dimensional reference template and the historical structure template, and determine whether the relative position characteristics between the corresponding connection blocks, turning blocks, main load-bearing blocks, and conventional structure blocks match. If they match, it is determined that the historical structure template matches the multi-dimensional reference template.

[0079] Among them, the method for determining whether the relative position characteristics match includes:

[0080] Compare the block relative position expression models of the multi-dimensional reference template and the historical structure template. For each block reference mapping point, randomly select several adjacent block reference mapping points, calculate the mapping point distance between the block reference mapping point and the adjacent block reference mapping points, and calculate the average value of the mapping point distances, denoted as the average mapping point distance, and associate the average mapping point distance with the block reference mapping point.

[0081] Determine the difference distance of each pair of corresponding block reference mapping points, and determine the sub-matching parameter corresponding to the block reference mapping point based on the average mapping point distance corresponding to the block reference mapping point. If the sub-matching parameter is greater than or equal to the preset value, it is determined that the corresponding block reference mapping point matches, and based on the proportion of the matching block reference mapping points, determine the matching parameter of the relative position characteristics between the multi-dimensional reference template and the historical structure template. If the matching parameter is greater than or equal to the preset value, it is determined that the relative position characteristics match.

[0082] The expression for calculating the matching parameter of the relative position feature is as follows: .

[0083] Where W is the matching parameter, J is the matching parameter conversion coefficient, is the sub-matching parameter judgment function of the i-th pair of corresponding block mapping points. If the sub-matching parameter is greater than or equal to the preset value, then output 1, otherwise output 0. n is the total number of pairs of all corresponding block mapping points, is the number of block mapping points of the block relative position expression model corresponding to the multi-dimensional reference template, is the number of block mapping points of the block relative position expression model corresponding to the historical structure template. c is the block mapping point matching influence adjustment constant, and R is the block mapping point matching influence adjustment coefficient.

[0084] Where the expression for calculating the sub-matching parameter is as follows: .

[0085] Where is the sub-matching parameter of the i-th pair of corresponding block reference mapping points, is the preset maximum sub-matching parameter, is the difference distance of the i-th pair of corresponding block reference mapping points, is the average distance between mapping points corresponding to the i-th pair of corresponding block reference mapping points, is the preset standard distance between mapping points, and f is the mapping point distance influence adjustment coefficient.

[0086] Step S200: Based on the structure state change data, classify the historical structure parameter groups to form several historical structure parameter sets. Each historical structure parameter set is randomly compared according to its own structure state change data and associated according to the equivalence degree, and several historical structure parameter group clusters are formed in the historical structure parameter sets.

[0087] After obtaining the historical structure parameter groups based on the geometric information multi-dimensional data reference template and their structure state data changing with time, the next step is to classify and analyze these data. The key is to compare the structure state changes of different parameter groups in the time series, find parameter groups with similar change patterns or trends, and classify them into the same set. By further comparing and analyzing within these sets, the internal relationships between the structure state changes and time, environmental factors, load conditions, etc. can be revealed, and several historical structure parameter group clusters with time evolution characteristics are formed.

[0088] In some embodiments disclosed by the present invention, the method for classifying historical structure parameter groups based on structure state change data includes:

[0089] Step S201: Analyze the structure state change data corresponding to each historical structure parameter, determine a number of structure factor state change data, analyze each structure factor state change data, determine the structure factor state parameters at different time nodes, construct a structure factor state parameter change curve, and record the combination of structure factor state parameter change curves belonging to the same structure state change data as a structure factor state parameter change curve group.

[0090] Step S202: Classify the structure factor state parameter change curve groups according to the similarity relationship between the structure factor state parameter change curve groups, and classify the historical structure parameter groups corresponding to the structure factor state parameter change curve groups according to the classification of the structure factor state parameter change curve groups.

[0091] Among them, the method for determining the similarity relationship between the structure factor state parameter change curve groups includes:

[0092] Align the corresponding structure factor state parameter change curves, compare the curvature integral values of several preset equivalent time intervals. If the difference amount of the curvature integral values is within the preset interval, it is determined that the corresponding structure factor state parameter change curves are similar to each other.

[0093] Determine the number of similar curves of the structure factor state parameter change curves with a mutual similarity relationship. If the ratio of the number of similar curves to the total number of all structure factor state parameter change curves is greater than or equal to the preset value, it is determined that there is a similarity relationship between the structure factor state change curve groups.

[0094] In some embodiments disclosed by the present invention, the method for determining the equivalent degree between the structure state change data includes:

[0095] Step S203: Compare the structure factor state parameter change curve groups corresponding to the structure state change data. There are several curve probe points set for the structure factor state parameter change curves. Calculate the curve probe point difference value corresponding to each curve probe point respectively. If the curve probe point difference value is less than or equal to the preset value, it is determined that the corresponding curve probe points are equivalent and recorded as equivalent curve probe points.

[0096] Step S204: If the ratio of the number of equivalent curve probe points belonging to the same structural factor state parameter change curve to the number of all curve probe points belongs to the first preset ratio interval, mark the first sub-equivalent degree for the corresponding structural factor state parameter curve; if the ratio of the number of probe points belongs to the second preset ratio interval, mark the second sub-equivalent degree for the corresponding structural factor state parameter curve; if the ratio of the number of probe points belongs to the third preset ratio interval, mark the third sub-equivalent degree for the corresponding structural factor state parameter curve;..., if the ratio of the number of probe points belongs to the nth preset ratio interval, mark the nth sub-equivalent degree for the corresponding structural factor state parameter curve.

[0097] Step S205: Accumulate and sum up the sub-equivalent degrees corresponding to all structural factor state parameter change curves to obtain the equivalent degree between structural state change data.

[0098] Step S300: Determine the priority reference order of each parameter group based on the number of parameter groups in the historical structural parameter group cluster in the historical structural parameters.

[0099] After determining the historical structural parameter group cluster based on the structural state change data, according to the number of parameter groups in the cluster and the universality and typicality of the structural state change patterns they represent, determine the priority reference order of each parameter group. The purpose of this step is to be able to give priority to considering those parameter groups that are more representative and can better reflect the law of structural state change over time in subsequent comparison and analysis, thereby improving the accuracy and efficiency of the analysis.

[0100] Step S400: Based on the multi-dimensional data reference template, obtain real-time structural parameters in real time to form a real-time structural parameter group, and based on the priority reference order of the historical structural parameter groups, compare and match the real-time structural parameter group with the historical structural parameter groups in sequence. Based on the matching results, select the most matching historical structural parameter group, and evaluate the reliability of the entity structure corresponding to the real-time structural parameter group based on the structural state change data associated with the most matching historical structural parameter group.

[0101] Use the geometric information multi-dimensional data reference template to obtain the geometric parameter data of the current entity structure in real time, and combine real-time monitoring technology to obtain the real-time state data of the structure to form a real-time structural parameter group. Then, according to the previously determined priority reference order, compare and match the real-time parameter group with the historical structural parameter groups with similar time evolution characteristics. By finding the historical parameter group whose structural state change in the time series is most consistent with the real-time parameter group, the reliability of the current entity structure can be dynamically evaluated using the associated historical structural state change data. This method combines the analogical reasoning of historical data and the monitoring analysis of real-time data, can more accurately reflect the change of structural state over time, and provides a scientific basis for the maintenance, repair and optimization of the structure.

[0102] In some embodiments disclosed by the present invention, the method for comparing and matching a real-time structure parameter group and a historical structure parameter group includes:

[0103] Step S401: Set attention weight coefficients for different structure parameters, and combine the structure parameter differences of each structure parameter between the structure parameter groups to determine the sub-structure difference parameters corresponding to the structure parameters. Based on the sub-structure difference parameters corresponding to all structure parameters, determine the structure difference parameter between the real-time structure parameter group and the historical structure parameter group.

[0104] Step S402: If the structure difference parameter is less than or equal to the preset value, determine that the historical structure parameter group is the most matching historical structure parameter group.

[0105] Among them, the expression for calculating the structure difference parameter is: .

[0106] Among them, Y is the structure difference parameter, is the sub-structure difference parameter of the x-th structure parameter, is the attention weight coefficient of the x-th structure parameter, and N is the number of structure parameters.

[0107] In some embodiments disclosed by the present invention, the method for evaluating the reliability of the entity structure corresponding to the real-time structure parameter group based on the structure state change data includes:

[0108] Step S403: Analyze the structure state change data to determine several structure factor state change data. Set several state change intervals for each structure factor state change data, and set corresponding reliability parameters for each state change interval. Based on the state change interval to which each structure factor state change data belongs, determine the reliability parameter corresponding to the structure factor state change data.

[0109] Step S404: Evaluate the reliability of the entity structure based on the reliability parameters corresponding to different structure factor state change data.

[0110] In some embodiments disclosed by the present invention, the present invention also discloses a structure reliability analysis system based on multi-dimensional data analysis, including:

[0111] The first module: Construct a multi-dimensional data reference template, and based on the multi-dimensional data reference template, analyze the historical structure reliability record data, extract several historical structure parameter groups, and the structure state change data corresponding to each historical structure parameter group;

[0112] A second module, configured to classify historical structure parameter groups based on structure state change data to form a number of historical structure parameter sets, and randomly compare each historical structure parameter set according to its respective structure state change data, and associate them according to the degree of equivalence, so as to form a number of historical structure parameter group clusters in the historical structure parameter sets;

[0113] A third module, configured to determine the priority reference order of each parameter group based on the number of parameter groups in the historical structure parameter group clusters;

[0114] A fourth module, configured to obtain real-time structure parameters in real time based on a multi-dimensional data reference template to form a real-time structure parameter group, and based on the priority reference order of the historical structure parameter groups, compare and match the real-time structure parameter group with the historical structure parameter groups in sequence, and based on the matching result, select the most matching historical structure parameter group, and evaluate the reliability of the entity structure corresponding to the real-time structure parameter group based on the structure state change data associated with the most matching historical structure parameter group.

[0115] The present invention discloses a structure reliability analysis method and system based on multi-dimensional data analysis, which relates to the technical field of structure reliability analysis. A multi-dimensional data reference template is constructed, and in-depth analysis is performed on historical structure reliability data to extract historical structure parameter groups and their corresponding state change data; the historical structure parameter groups are classified according to the structure state change data to form historical structure parameter sets, and further divided into historical structure parameter group clusters; real-time structure parameters are obtained in real time to form a real-time structure parameter group, and compared and matched with the historical parameter groups according to the priority reference order; by selecting the most matching historical parameter group and using the associated structure state change data, the reliability of the current entity structure is accurately evaluated. The above technical solution of the present invention realizes the reliability evaluation of the entity structure and provides data support for ensuring the safe operation of the entity structure.

[0116] Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented through hardware or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present invention.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions of the present invention or make equivalent replacements, and these modifications or equivalent replacements do not enable the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A structural reliability analysis method based on multidimensional data analysis, characterized in that: include: Constructing a multidimensional data reference template, and based on the multidimensional data reference template, analyzing the historical structural reliability record data, extracting a number of historical structural parameter groups, and structural state change data corresponding to each historical structural parameter group; Based on the structural state change data, the historical structural parameter groups are classified to form several historical structural parameter sets, and each historical structural parameter set is randomly compared according to its own structural state change data, and associated according to the degree of equality, forming several historical structural parameter group clusters in the historical structural parameter set; Based on the number of parameter groups of the historical structural parameter groups in the historical structural parameter group cluster, determining a priority reference order for each parameter group; Based on the multi-dimensional data reference template, real-time structural parameters are obtained in real time to form a real-time structural parameter group, and based on the priority reference sequence of the historical structural parameter group, the real-time structural parameter group is sequentially compared and matched with the historical structural parameter group, and based on the matching result, the most consistent historical structural parameter group is selected, and based on the structural state change data associated with the most consistent historical structural parameter group, the reliability of the physical structure corresponding to the real-time structural parameter group is evaluated; Based on the structural state change data, the methods for classifying the historical structural parameter groups include: Analyze the structural state change data corresponding to each historical structural parameter to determine a number of structural factor state change data, and analyze each structural factor state change data to determine the structural factor state parameters at different time nodes, and construct a structural factor state parameter change curve, and record the combination of structural factor state parameter change curves belonging to the same structural state change data as a structural factor state parameter change curve group; Methods for determining the degree of equality between structural state change data include: Compare the structure factor state parameter change curve group corresponding to the structure state change data, set a number of curve probe points for the structure factor state parameter change curve, calculate the curve probe point difference value corresponding to each curve probe point, and if the curve probe point difference value is less than or equal to the preset value, the corresponding curve probe points are deemed to be identical and recorded as identical curve probe points; If the ratio of the number of probe points of the identical curve probe points to all the curve probe points belonging to the same structural factor state parameter change curve belongs to the first preset ratio interval, then the first sub-equivalence degree is marked for the corresponding structural factor state parameter curve; if the ratio of the number of probe points belongs to the second preset ratio interval, then the second sub-equivalence degree is marked for the corresponding structural factor state parameter curve; if the ratio of the number of probe points belongs to the third preset ratio interval, then the third sub-equivalence degree is marked for the corresponding structural factor state parameter curve; ..., if the ratio of the number of probe points belongs to the nth preset ratio interval, then the nth sub-equivalence degree is marked for the corresponding structural factor state parameter curve; The sub-equivalence degrees corresponding to all the structural factor state parameter change curves are accumulated and summed to obtain the equivalence degree between the structural state change data; The method for evaluating the reliability of the physical structure corresponding to the real-time structural parameter group based on the structural state change data includes: Analyze the structural state change data to determine a number of structural factor state change data, set a number of state change intervals for each structural factor state change data, and set a corresponding reliability parameter for each state change interval, and determine the reliability parameter corresponding to the structural factor state change data based on the state change interval to which each structural factor state change data belongs; The reliability of the physical structure is evaluated based on the reliability parameters corresponding to the state change data of different structural factors.

2. The structural reliability analysis method based on multidimensional data analysis according to claim 1 is characterized in that: Methods for constructing a multidimensional data reference template include: Conduct three-dimensional structural analysis on the corresponding physical structure in the historical structural reliability record data, and determine the connection blocks, turning blocks, main load-bearing blocks and conventional structural blocks on the physical structure; Construct a three-dimensional analysis space, set a positioning coordinate system for the three-dimensional analysis space, configure the physical structure in the historical structural reliability record data in the positioning coordinate system, and configure the determined connection blocks, turning blocks, main load-bearing blocks and conventional structure blocks in the positioning coordinate system; The relative position characteristics between the connecting blocks, turning blocks, main load-bearing blocks and conventional structure blocks are recorded, and the historical structural parameters of each block are determined and recorded. The relative position characteristics are set on the positioning coordinate system to construct a multi-dimensional data reference template, in which each block is set with a number of historical structural parameters, and all historical structural parameters corresponding to all blocks are recorded as a historical structural parameter group; Among them, the connecting block is the block corresponding to the connecting part in the structure, the turning block is the block corresponding to the part where the shape or direction of the structure changes significantly, the main load-bearing block is the block corresponding to the part that bears the main load or stress, and the conventional structure block is the block corresponding to other parts except the above-mentioned special blocks.

3. The structural reliability analysis method based on multidimensional data analysis according to claim 2 is characterized in that: The expressions of the relative position characteristics between the connecting blocks, turning blocks, main load-bearing blocks and conventional structural blocks include: Determine the connection blocks, turning blocks, main load bearing blocks and conventional structure blocks respectively, and randomly select a number of block random mapping points for each block respectively, and calculate the average value of the block random mapping points of each block, which is recorded as the block reference mapping point; The block reference mapping points corresponding to each block are connected to form a block relative position expression model.

4. The structural reliability analysis method based on multidimensional data analysis according to claim 3 is characterized in that: Based on the multidimensional data reference template, the method for analyzing the historical structural reliability record data includes: Analyze the historical structure reliability record data to determine the historical structure template, historical structure parameter group and structure state change data corresponding to the historical monomer record data, and determine the historical structure parameter group and structure state change data to be called based on the matching relationship between the multidimensional data reference template and the historical structure template; The method for determining the matching relationship between the multi-dimensional reference template and the historical structure template includes: Compare the multi-dimensional reference template with the historical structure template to determine whether the relative position characteristics between the corresponding connecting blocks, turning blocks, main load-bearing blocks and conventional structure blocks are consistent. If they are consistent, it is determined that the historical structure template matches the multi-dimensional reference template; Among them, the method of determining whether the relative position features match includes: Compare the block relative position expression models of the multi-dimensional reference template and the historical structure template, randomly select a number of adjacent block reference mapping points for each block reference mapping point, calculate the distance between the block reference mapping point and the adjacent block reference mapping points, and calculate the average value of the distance between the mapping points, recorded as the average distance between the mapping points, and associate the average distance between the mapping points with the block reference mapping point; Determine the difference distance of each relative block reference mapping point, and determine the sub-matching parameter corresponding to the block reference mapping point based on the average distance between mapping points corresponding to the block reference mapping point. If the sub-matching parameter is greater than or equal to a preset value, it is determined that the corresponding block reference mapping point is matched, and based on the proportion of the matched block reference mapping points, determine the matching parameter of the relative position feature of the multi-dimensional reference template and the historical joint structure template. If the matching parameter is greater than or equal to the preset value, it is determined that the relative position feature is matched; The expression for calculating the matching parameter of the relative position feature is: ； in, For the matching parameters, is the matching parameter conversion coefficient, For the The sub-matching parameter judgment function of the corresponding block mapping point, if the sub-matching parameter is greater than or equal to the preset value, then Output 1, otherwise output 0. is the total number of pairs of all corresponding block mapping points, is the number of block mapping points of the block relative position expression model corresponding to the multi-dimensional reference template, is the number of block mapping points of the block relative position expression model corresponding to the historical structure template, Adjust constants for block mapping point matching effects, The adjustment coefficient for the matching impact of block mapping points; Among them, the expression for calculating the sub-matching parameter is: ; in, For the Sub-matching parameters for the corresponding block reference mapping points, is the preset maximum sub-matching parameter, For the The difference distance of the corresponding block reference mapping point, For the The average distance between mapping points corresponding to the corresponding block reference mapping points, is the preset standard mapping point distance, It is the distance adjustment factor between mapping points.

5. The structural reliability analysis method based on multidimensional data analysis according to claim 1 is characterized in that: Based on the structural state change data, the methods for classifying the historical structural parameter groups include: According to the similarity relationship between the structure factor state parameter change curve groups, the structure factor state parameter change curve groups are classified, and according to the classification of the structure factor state parameter change curve groups, the historical structure parameter groups corresponding to the structure factor state parameter change curve groups are classified; The method for determining the similarity relationship between the structure factor state parameter change curve groups includes: Aligning the corresponding structure factor state parameter change curves, and comparing the curvature integral values ​​of a number of preset equivalent time segments, if the difference in the curvature integral value is within a preset interval, it is determined that the corresponding structure factor state parameter change curves are similar to each other; Determine the number of similar curves of the structural factor state parameter change curves that have a mutual similarity relationship. If the ratio of the number of similar curves to the total number of curves of all structural factor state parameter change curves is greater than or equal to a preset value, it is determined that there is a similarity relationship between the structural factor state change curve groups.

6. The structural reliability analysis method based on multidimensional data analysis according to claim 1 is characterized in that: The method for comparing and matching the real-time structural parameter group and the historical structural parameter group includes: A focus weight coefficient is set for different structural parameters, and the sub-structural difference parameter of the corresponding structural parameter is determined in combination with the structural parameter difference of each structural parameter between the structural parameter groups. Based on the sub-structural difference parameters corresponding to all structural parameters, the structural difference parameter between the real-time structural parameter group and the historical structural parameter group is determined; If the structural difference parameter is less than or equal to the preset value, the historical structural parameter group is determined to be the most consistent historical structural parameter group; Among them, the expression for calculating the structural difference parameter is: ； in, is the structural difference parameter, For the The substructure difference parameter of the structural parameters, For the The attention weight coefficient of the structural parameters is is the number of structural parameters.

7. The structural reliability analysis system based on multidimensional data analysis is characterized by: A method for performing a structural reliability analysis based on multidimensional data analysis according to any one of claims 1 to 6, comprising: The first module constructs a multidimensional data reference template, and based on the multidimensional data reference template, analyzes the historical structural reliability record data, extracts several historical structural parameter groups, and the structural state change data corresponding to each historical structural parameter group; The second module is used to classify the historical structural parameter groups based on the structural state change data to form a number of historical structural parameter sets, and each historical structural parameter set is randomly compared according to its own structural state change data, and is associated according to the degree of equality to form a number of historical structural parameter group clusters in the historical structural parameter set; The third module is used to determine the priority reference order of each parameter group based on the number of parameter groups of the historical structure parameter groups in the historical structure parameter group cluster; The fourth module is used to obtain real-time structural parameters in real time based on the multidimensional data reference template to form a real-time structural parameter group, and based on the priority reference sequence of the historical structural parameter group, the real-time structural parameter group is compared and matched with the historical structural parameter group in turn, and based on the matching result, the most consistent historical structural parameter group is selected, and based on the structural state change data associated with the most consistent historical structural parameter group, the reliability of the physical structure corresponding to the real-time structural parameter group is evaluated.

Images