CN110969370A - Quality risk analysis method for building structural member - Google Patents
Quality risk analysis method for building structural member Download PDFInfo
- Publication number
- CN110969370A CN110969370A CN201911324065.0A CN201911324065A CN110969370A CN 110969370 A CN110969370 A CN 110969370A CN 201911324065 A CN201911324065 A CN 201911324065A CN 110969370 A CN110969370 A CN 110969370A
- Authority
- CN
- China
- Prior art keywords
- data
- ica
- structural member
- dimension
- regression
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012502 risk assessment Methods 0.000 title claims abstract description 12
- 238000012880 independent component analysis Methods 0.000 claims abstract description 51
- 230000009467 reduction Effects 0.000 claims abstract description 22
- 238000000342 Monte Carlo simulation Methods 0.000 claims abstract description 18
- 238000001514 detection method Methods 0.000 claims abstract description 16
- 238000004364 calculation method Methods 0.000 claims abstract description 14
- 238000012544 monitoring process Methods 0.000 claims abstract description 14
- 230000003044 adaptive effect Effects 0.000 claims abstract description 11
- 238000000611 regression analysis Methods 0.000 claims abstract description 11
- 238000011156 evaluation Methods 0.000 claims abstract description 10
- 238000002790 cross-validation Methods 0.000 claims description 14
- 230000001419 dependent effect Effects 0.000 claims description 11
- 238000012360 testing method Methods 0.000 claims description 10
- 238000006073 displacement reaction Methods 0.000 claims description 9
- 229910000831 Steel Inorganic materials 0.000 claims description 4
- 239000010959 steel Substances 0.000 claims description 4
- 238000007689 inspection Methods 0.000 claims description 3
- 238000011835 investigation Methods 0.000 claims description 2
- 239000000463 material Substances 0.000 claims description 2
- 238000006386 neutralization reaction Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000004422 calculation algorithm Methods 0.000 description 9
- 230000007547 defect Effects 0.000 description 8
- 238000007667 floating Methods 0.000 description 8
- 238000010276 construction Methods 0.000 description 6
- 230000000875 corresponding effect Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 230000001276 controlling effect Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000003908 quality control method Methods 0.000 description 3
- 238000004445 quantitative analysis Methods 0.000 description 3
- 238000012954 risk control Methods 0.000 description 3
- 238000013178 mathematical model Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000011435 rock Substances 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 241001261505 Undaria Species 0.000 description 1
- 238000004378 air conditioning Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000007586 pull-out test Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 230000003014 reinforcing effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000010117 shenhua Substances 0.000 description 1
- 238000004154 testing of material Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
- 238000004078 waterproofing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2134—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on separation criteria, e.g. independent component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
A method of quality risk analysis of an architectural structural member, comprising: storing engineering detection and monitoring data of the structural member into a computer readable format; carrying out dimensionality reduction trial calculation on independent variable data by adopting an Independent Component Analysis (ICA) method; performing regression analysis on the m-1 data sets by adopting a multivariate adaptive regression MARS; according to the evaluation result, selecting an optimal ICA dimension reduction data set and an ICA dimension reduction dimension n corresponding to the data set; obtaining a predicted final model of the resistance of the structural member by regression of a multivariate adaptive regression MARS; based on the predicted final model R and the load; and simulating the model by adopting a Monte Carlo method so as to obtain the transcendental probability or the failure probability. The invention provides the risk analysis and prediction conclusion with failure probability or overtaking probability as output, which is beneficial to getting through the gap between the financial insurance industry and the engineering industry.
Description
Technical Field
The invention relates to the field of buildings, in particular to a quality risk analysis method for a building structural member.
Background
The existing engineering quality control method generally judges whether engineering can be accepted or not and whether the engineering meets a qualified line of standard requirements or not through engineering detection, and cannot provide a quantitative method which gives probability based on performance. The traditional project acceptance and qualification method is not enough for the project quality defect insurance risk control mechanism which is greatly promoted in China in recent years (the failure probability or the overtaking probability can not be predicted generally), but the project quality probability prediction method based on big data can get through the keys of insurance risk control, rate determination and even actuarial.
Construction project quality latent defects insurance (IDI) is applied by a construction unit, and an insurance company applies insurance for reimbursing obligations for damages of an applied building caused by the latent defects of the construction quality within an insurance range and an insurance limit according to insurance clauses. IDI belongs to the ship, and research and practice have been carried out very early in countries of the european union, such as finland, france, italy and spain, where IDI was carried out as a mandatory insurance. The implementation detailed rule time lines of the project quality defect insurance (IDI) are provided in Beijing, Shanghai and Shenzhen, and the basic insurance coverage of the IDI is specified in the temporary insurance management method of the project quality potential defect insurance of Beijing city, which is provided in Beijing in 2019, and is foundation and major structure project (10 years), heat preservation and waterproof project (5 years):
the defects of foundation and main structure engineering include: 1. collapse wholly or partially; 2. the foundation generates uneven settlement exceeding the design specification; 3. cracks, deformation, damage and fracture affecting the structure safety appear on the foundation and the main structure part; 4. cracks, deformation, damage and fracture occur to overhanging components such as balconies, rain canopies, cornices, air-conditioning boards and the like, which influence the use safety; 5. quality defects of use safety influenced by falling and collapse of the outer wall surface; 6. and potential defects of engineering quality influencing the structural safety appear on other foundation foundations and main structure parts.
Insulation and waterproofing engineering drawbacks include: 1. the insulating layer of the enclosure structure is damaged and falls off; 2. water leakage prevention in underground, roof and bathroom; 3. leakage from the exterior wall (including the junction of the exterior window and the exterior wall); 4. other parts with waterproof requirements leak.
The quality of the engineering quality can be evaluated by a qualitative or quantitative method. Within the scope of the quantitative method, the probability method based on monte carlo simulation can intuitively give the failure probability or the exceeding probability of the building structural member, and the probability level directly reflects the quality of the building structural member, but the method needs to be applied to know the extreme state equation of the structural member, and the extreme state equation Z can be expressed by resistance R and load Q in mathematics as follows:
Z＝g(R，Q)＝R-Q
when (1) Z is more than 0, the building structural member meets the functional requirement; (2) z is less than 0, the building structural member fails or the load level Q exceeds the preset resistance performance index R warning value; (3) and Z is 0, and the building structural member is in a critical state. It must be noted that the resistance R and the load Q here do not only include the variables of force, stress, but also the variables of interest in other construction works, such as deformation, settlement, seepage, etc.
In practice, however, the extreme state equations of many architectural structural members are difficult to obtain, particularly those involving subterranean structural members, because uncertainties in the properties of the ground-based rock-soil mass and uncertainties in the boundary conditions result in structural members that have no deterministic mathematical and mechanical models; the limit state equation of some upper structural members can be obtained, but the practical boundary conditions are too complex, so that the numerical simulation calculation cost (economic cost) is higher, and the economic benefit is not obtained in the practical engineering in consideration.
The invention provides a method for controlling and checking engineering quality process which is carried out in China construction engineering field legal by engineering detection or monitoring of building structural members. Whether the existing building structural member has a known extreme state equation or not, the extreme state equation based on the detection or monitoring data can be established by the method provided by the invention, and the failure probability or the exceeding probability of the building structural member is further calculated by Monte Carlo simulation.
In the prior art, the risk analysis with probability output mainly uses a monte carlo method in the construction field and is concentrated in bridge and tunnel engineering, for example, a method for establishing a limit state equation by combining a finite element model and a response surface based on polynomial expansion is proposed in the chinese patent invention CN 107844651 a.
Disclosure of Invention
The invention provides a quality risk analysis method of a building structural member, which aims to solve at least one technical problem.
To solve the above problems, as an aspect of the present invention, there is provided a quality risk analysis method of a building structural member, including:
step one, storing engineering detection and monitoring data of a structural member into a computer readable format, wherein the data comprises a plurality of variables, one of the variables is a dependent variable, the other variables are independent variables, and the dependent variable is the resistance of the structural member;
step two, carrying out dimensionality reduction trial calculation on the independent variable data by adopting an Independent Component Analysis (ICA) method, wherein the reduced dimensionality n is less than or equal to the dimensionality m of the input data, and the dimensionality n is traversed from 1 dimension to m-1 dimension, so that m-1 data sets are obtained in the step;
step three, performing regression analysis on the m-1 data sets by adopting a multivariate adaptive regression MARS, evaluating the quality of the regression analysis result by adopting a cross validation method, and respectively giving evaluation result indexes of mean square error MSE, GCV and R to the m-1 data sets2As shown in the following formula:
wherein, yiIs the predicted value of the i-th dependent variable, YiIs the measured value of the ith dependent variable, d is 3, M is the number of basis functions in the multivariate adaptive regression MARS, N is the number of data,
the average value of the variables;
selecting an optimal ICA dimension reduction data set and an ICA dimension reduction n corresponding to the data set according to the evaluation result, wherein the selection standard is the lowest MSE, the lowest GCV and the highest R2 or any combination of the three;
step five, using all data sets of the ICA dimension reduction dimension n selected in the step four and the corresponding dimension reduction for the regression of the multi-element adaptive regression MARS to obtain a predicted final model R of the resistance of the structural member;
step six, based on the predicted final model R of the resistance of the structural member and the load Q, giving a limit state equation Z which is R-Q, wherein Q can be a limit value given according to national, industrial or local standards, a trigger value in insurance clauses or a value set by a user;
and seventhly, simulating the model by adopting a Monte Carlo method, thereby obtaining the exceeding probability or the failure probability of the structural member.
Preferably, the structural members include beams, columns, plates, piles, anchors, and the like.
Preferably, the engineering detection and monitoring data includes: load-displacement data or stress-strain data of a field test, related data (physical and mechanical parameters of a geotechnical layer) in a geotechnical engineering investigation report, and material test data of concrete, steel bars and the like.
Preferably, the step four neutralization selection criteria depend on the purpose of the person using the data, and ideally the three indexes point to the same ICA-reduced dataset, MSE is applied if absolute error priority cannot be met, R is applied if capture of data laws is prioritized2GCV should be used if the main objective is to consider reducing the overfitting.
The invention has the advantages that:
1) the invention is purely data-driven and is suitable for various building structural members.
2) The engineering detection and data based on the method generally come from quality control projects which are required to be finished in the engineering process and acceptance link, and the method can be used by separately and additionally collecting detection or monitoring data.
3) The invention provides the risk analysis and prediction conclusion with failure probability or overtaking probability as output, which is beneficial to getting through the gap between the financial insurance industry and the engineering industry.
Drawings
FIG. 1 schematically illustrates a flow chart of the present invention;
fig. 2 schematically shows a schematic view of an anti-floating anchor pull-out test apparatus;
fig. 3 schematically illustrates a modeling and model validation process of a bolt load-displacement prediction model;
FIG. 4 schematically shows the behavior of MSE and GCV in different ICA dimensions;
FIG. 5 schematically shows R2Performance under different ICA dimensions;
FIG. 6 schematically illustrates a comparison of measured deformation values of the anchor with predicted values of the ICA-MARS model;
fig. 7 schematically shows a bolt load-displacement curve: comparison of the measured values with the predicted values of the ICA-MARS model.
Detailed Description
The embodiments of the invention will be described in detail below with reference to the drawings, but the invention can be implemented in many different ways as defined and covered by the claims.
The invention provides a quality risk analysis method of a building structural member, which is essentially a mathematical model for evaluating the quality of the building structural member, which is established based on building engineering detection and monitoring data, and the mathematical model comprises the key technology, specific implementation algorithm and steps of the independent innovation of the invention. The input to the model is engineering inspection and monitoring data and the output of the model is (override or failure) probability, the model being purely data driven, i.e. without modifying the model for different building structure element types.
The purpose of the invention is: based on the legally necessary engineering detection or monitoring data in engineering, the invention adopts the self-innovative machine learning method to obtain the extreme state equation of the building structural member and the equation variables which are independent of each other in statistics. On the basis, the failure probability or the exceeding probability of the building structural member is obtained by adopting a universal Monte Carlo simulation method, and an intuitive basis is provided for engineering insurance risk control.
From engineering detection or monitoring data to failure probability or transcendental probability, the method based on combination of machine learning and Monte Carlo simulation is ICA-MARS-MC, wherein ICA represents an independent principal component analysis algorithm, MARS refers to a multivariate self-adaptive regression spline, MC refers to Monte Carlo simulation, and ICA-MARS-MC forms an algorithm chain by the three algorithms to make up for the deficiency of each other and improve the prediction accuracy. The specific procedure for ICA-MARS-MC can be described as follows:
the method comprises the following steps: the engineering detection and monitoring data of structural members such as beams, columns, plates, piles, anchor rods and the like, such as load-displacement data or stress-strain data of field tests, related data (physical and mechanical parameters of rock-soil layers), concrete, steel bars and other material testing data in geotechnical engineering survey reports are automatically or manually stored into a computer-readable format, such as an excel spreadsheet, a csv format and the like, generally, the data have multiple dimensions (variables), wherein one variable is the resistance of the structural member, the variable is used as a dependent variable (a predicted object) here, and the other variables are independent variables;
step two: performing dimensionality reduction trial calculation on the independent variable data input in the step one by adopting an Independent Component Analysis (ICA) method, wherein the reduced dimensionality n is less than or equal to the dimensionality m of the input data, and the dimensionality n is traversed from 1 dimension to m-1 dimension, so that m-1 data sets are obtained in the step;
step three: performing regression analysis on the m-1 data sets obtained in the step two by adopting a multivariate adaptive regression MARS, evaluating the quality of the regression analysis result by adopting a cross validation method, and respectively giving evaluation result indexes of Mean Square Error (MSE), GCV and R to the m-1 data sets2As shown in the following formula:
in the formula yiIs the predicted value of the i-th dependent variable, YiIs the measured value of the ith dependent variable, d is 3, M is the number of basis functions in the multivariate adaptive regression MARS, N is the number of data,
is the average of the variables.
The mean square error MSE is the mean value of the square sum of errors of corresponding points of the predicted data and the original data; on the basis that MSE is used as a numerator, the generalized cross validation mean square error GCV describes the complexity of a model in a denominator, so that indexes of the generalization capability and precision of the model are comprehensively considered; determining the coefficient R2The method is an index of a common evaluation model, and a molecular part represents the sum of the square differences of a true value and a predicted value, and is similar to Mean Square Error (MSE); the denominator part represents the sum of the squared differences of the true value and the mean, R2The value range is [0,1 ]]In general, a closer to 1 indicates a better model effect
Step four: according to the evaluation results in the three steps, an optimal ICA dimension-reduced data set and an ICA dimension-reduced dimension n corresponding to the data set are selected, the selection criteria are the lowest MSE, the lowest GCV and the highest R2 or any combination of the three, and the data set is determined according to the purpose of a person using the data, if the three indexes point to the same ICA dimension-reduced data set in an ideal state, the MSE is applied when the absolute error is considered to be preferential, if the absolute error is not satisfied, the R2 is applied when the capture priority of the data law is considered, and if the reduction of overfitting is considered as a main target, the GCV is applied.
Step five: using all the ICA dimension reduction n selected in the fourth step and all the data sets (noted all, not the cross validation method in the third step) corresponding to the dimension reduction for the regression of the multi-element adaptive regression MARS to obtain a predicted final model R of the resistance of the structural member;
step six: based on the predicted final model R of the resistance of the structural member and the load Q, which are given in the step five, a limit state equation Z is given as R-Q, wherein Q can be a limit value given according to national, industrial or local standards, a trigger value in insurance clauses or a value set by a user;
step seven: and (3) simulating the model by adopting a Monte Carlo method, so as to obtain the transcending probability or the failure probability of the structural member (namely, the statistical ratio of Z <0 in the Monte Carlo simulation is the transcending probability or the failure probability of the structural member).
The method based on the combination of machine learning and Monte Carlo simulation has the advantages that:
(1) by using ICA as data preprocessing, the data dimensionality reduction effect is realized, and the prediction precision can be effectively improved by obtaining statistic mutually independent recessive dimensionality through ICA analysis compared with the independent regression analysis;
(2) the variables obtained by ICA transformation are mutually independent, which provides a basis for generating mutually independent samples by random sampling when the MC method is used, and avoids the uncertainty of an original data set caused by sample sparseness on correlation coefficient matrix fitting;
(3) the explicit prediction model provided by the MARS not only facilitates more intuitive understanding of engineers, but also greatly reduces the calculation cost compared with numerical calculation of a finite element method and the like, and provides possibility for obviously increasing the number of samples when the MC is used subsequently so as to improve the calculation precision.
The key and innovation points of the invention are as follows:
1) in the first step of the method, the Independent Component Analysis (ICA) algorithm is used for dimensionality reduction to obtain a group of dimensionality reduced variables which are statistically independent to each other.
2) The invention adopts cross validation after MARS regression analysis to select dimension reduction of ICA algorithm.
3) The resistance prediction model R of the building structural member established by the invention is established by MARS regression analysis after the dimensionality reduction of the ICA is determined by cross validation, namely the input of the step of the MARS regression analysis is the output of the ICA after the dimensionality reduction.
3) The invention uses ICA in the first step, and random sampling in the MC Monte Carlo simulation in the last step does not consider the correlation among variables, because the variables are not statistically correlated after ICA transformation.
The invention has the advantages that:
1) the invention is purely data-driven and is suitable for various building structural members.
2) The engineering detection and data based on the method generally come from quality control projects which are required to be finished in the engineering process and acceptance link, and the method can be used by separately and additionally collecting detection or monitoring data.
3) The invention provides the risk analysis and prediction conclusion with failure probability or overtaking probability as output, which is beneficial to getting through the gap between the financial insurance industry and the engineering industry.
Example (b): deformation prediction and failure probability evaluation case for engineering anti-floating anchor rod
The embodiment is a sampling drawing test carried out on the anti-floating anchor rod engineering in the engineering inspection and acceptance stage of the anti-floating anchor rod of a typical building underground structural member. In the test, a core-through jack and an anchorage device are used as drawing devices, and the counter force is transmitted to the ground through a steel plate. During the test, the hydraulic jack is adopted to load and unload loads in a grading way, the pressure gauge on the manual oil pump is used for controlling the loading amount of each grade, and the displacement sensor is used for measuring the displacement when the anchor head is pulled out during the loading.
Overview of engineering
The project is located in Shenzhen Shenhua region with floor area of about 54607.07m2The bottom of the tower is provided with 5 commercial undaria buildings, and the upper part of the tower is provided with 6 super high-rise towers (1 office building and 5 commercial and residential buildings); the building comprises an office building (44F)250m, 4 seats (48+4F)165m and 1 seat (29+4F)109.9 m in a commercial and residential building, and is buried at the depth of about 18m in an underground 4-layer building. The building structure form is as follows: the main building adopts a shear wall structure; the skirt room adopts a frame structure, and the basement adopts a plate column structure. The design grade of the engineering foundation is grade A, and the design grade of the building foundation pile is grade A. Designing a basic form: pile foundation, pier foundation, natural rock foundation and anti-floating anchor rod. For the anti-floating anchor rod, the anti-floating anchor rod is required to be a rock anchor rod, the rock-entering depth is not less than 3.5m, the pore-forming diameter is 180mm, the number of reinforcing bars is 3, the HRB400 is 32mm, and the characteristic value of the uplift bearing capacity is 450 kN.
TABLE 1 mechanical parameter suggestion table for natural foundation of engineering
Establishment of two, ICA-MARS-MC model
Determination of ICA dimensions and model validation
A drawing test is carried out on 87 anti-floating anchor rods aiming at the detection acceptance of the engineering to obtain 783 groups of test data, and according to a data division method, a modeling and model verification process shown in figure 3 is carried out. Where ICA and MARS are both purely data-driven algorithms, but the dimensionality of the output of the ICA algorithm is typically determined by an artificially set criterion, how are the dimensions of the ICA output determined in this project?
The project proposes that an optimal ICA dimension is selected by adopting a cross validation method, namely three model evaluation indexes of 5-fold cross validation average MSE, GCV and R2 under different ICA dimensions are calculated by a 5-fold cross validation method, so that the selection of the ICA dimension is completed. As shown in FIGS. 4 and 5, when the ICA dimension is 11, the 5-fold cross-validation model givesHas the minimum mean value of MSE and GCV, R2The highest, and therefore the comparison of the average results of cross-validation with such trial calculations controlling only one variable of the ICA dimension can clearly indicate that the model is optimal when the ICA dimension is 11.
While the dimension of ICA is determined, the model is verified by a 5-fold cross validation method, and the MSE is 1.96, GCV is 2.89, and R is known as the result2This accuracy should be acceptable for practical geotechnical issues, 0.92.
2. Prediction model
Through the above cross-validation modeling process, the model of ICA-MARS (ICA dimension 11) can be identified as the model used for final prediction, so all known data will be used for the training of the model to obtain the parameters of the model to build the prediction model with the best accuracy. The final prediction model is shown in table 2 with the prediction models given by ICAi (i ═ 0 to 10) of 11 ICA variables, and the evaluation index calculation results of the models are as follows: MSE 1.8159, GCV 2.6202, R2 0.9233. Fig. 6 and 7 respectively show a comparison graph of the measured deformation value of the anchor rod and a load-displacement curve, and it can be found that the model of ICA-MARS (ICA dimension is 11) gives better effect from the prediction of the deformation value and the simulation of the load-displacement curve form.
TABLE 2 ICA-MARS prediction model
| Basis function BF | Coefficient βm | Basis function BF | Coefficient βm |
| Intercept of a beam | -164.95 | h(-0.11089-ICA3) | 362.45 |
| h(ICA10-0.0105835) | -3600.26 | h(ICA7+0.0742366) | 69.1896 |
| h(0.0105835-ICA10) | 4288.52 | h(ICA7+0.219545) | 162.601 |
| h(ICA2-0.0175126) | -469.376 | h(-0.219545-ICA7) | -319.813 |
| h(0.0175126-V2) | 410.445 | h(0.0379017-ICA10) | -188.226 |
| h(0.068245-ICA10) | -1029.26 | h(0.220921-ICA2) | 182.498 |
| h(ICA10-0.051618) | 486.973 | h(0.0727203-ICA10) | 442.368 |
| h(-0.0289058-ICA4) | -113.691 | h(ICA0-0.075522) | -77.0152 |
| h(0.0603567-ICA4) | 80.2604 | h(ICA4-0.0807966) | 97.4059 |
| h(0.0445167-ICA0) | 174.393 | h(-0.22519-ICA6) | 77.3032 |
| h(ICA5+0.0901701) | 14621.9 | h(ICA5+0.0208572) | 225.696 |
| h(-0.0901701-ICA5) | -1556.77 | h(ICA5-0.0086334) | -131.23 |
| h(ICA5+0.0922799) | -14644.2 | h(ICA5-0.0580034) | 45.5171 |
| h(ICA10+0.0154523) | -513.484 | h(ICA9+0.151809) | -349.425 |
| h(-0.0788279-ICA0) | -155.33 | h(ICA10-0.119567) | -172.238 |
| h(0.0247939-ICA7) | 126.578 |
3. Monte Carlo method for calculating failure probability
The use of the previous ICA algorithm greatly facilitates the sampling process when the monte carlo method is applied, because 11 variables output by the ICA are statistically independent from each other, so that the relation between different variables, i.e. the correlation coefficient is 0, is not considered when generating random numbers in the monte carlo simulation, which technically solves an important problem: when the sample size is small and the data distribution is sparse, the real correlation coefficient or covariance matrix among the fitted variables of the sample may have large errors, and if the variables are statistically independent, namely the correlation coefficient is 0, the errors can be avoided when random numbers are generated in the Monte Carlo simulation.
Random () of a Python basic function is adopted to generate random numbers in the project, and due to the high efficiency of computer processing ICA-MARS explicit model, namely, table 4.2, the random sample numbers adopted in each calculation are millions of times, the obtained calculation results are shown in table 3, and the variation coefficient corresponding to the transcendental probability is extremely less than 0.05%. The average 32s calculation time also meets the efficiency requirements in practical engineering.
TABLE 3 results of Monte Carlo simulation calculations
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (4)
1. A method of quality risk analysis of an architectural structural member, comprising:
step one, storing engineering detection and monitoring data of a structural member into a computer readable format, wherein the data comprises a plurality of variables, one of the variables is a dependent variable, the other variables are independent variables, and the dependent variable is the resistance of the structural member;
step two, carrying out dimensionality reduction trial calculation on the independent variable data by adopting an Independent Component Analysis (ICA) method, wherein the reduced dimensionality n is less than or equal to the dimensionality m of the input data, and the dimensionality n is traversed from 1 dimension to m-1 dimension, so that m-1 data sets are obtained in the step;
step three, performing regression analysis on the m-1 data sets by adopting a multivariate adaptive regression MARS, evaluating the quality of the regression analysis result by adopting a cross validation method, and respectively giving evaluation result indexes of mean square error MSE, GCV and R to the m-1 data sets2As shown in the following formula:
wherein, yiIs the predicted value of the i-th dependent variable, YiIs the measured value of the ith dependent variable, d is 3, and M is in the multivariate adaptive regression MARSN is the number of data,
the average value of the variables;
selecting an optimal ICA dimension reduction data set and an ICA dimension reduction n corresponding to the data set according to the evaluation result, wherein the selection standard is the lowest MSE, the lowest GCV and the highest R2 or any combination of the three;
step five, using all data sets of the ICA dimension reduction dimension n selected in the step four and the corresponding dimension reduction for the regression of the multi-element adaptive regression MARS to obtain a predicted final model R of the resistance of the structural member;
step six, based on the predicted final model R of the resistance of the structural member and the load Q, giving a limit state equation Z which is R-Q, wherein Q can be a limit value given according to national, industrial or local standards, a trigger value in insurance clauses or a value set by a user;
and seventhly, simulating the model by adopting a Monte Carlo method, thereby obtaining the exceeding probability or the failure probability of the structural member.
2. The method of claim 1, wherein the structural members comprise beams, columns, plates, piles, anchors, and the like.
3. The method of claim, wherein said engineering inspection, monitoring data comprises: load-displacement data or stress-strain data of a field test, related data (physical and mechanical parameters of a geotechnical layer) in a geotechnical engineering investigation report, and material test data of concrete, steel bars and the like.
4. The method of claim 1, wherein the step four neutralization selection criteria depends on the purpose of the user, and ideally the three indicators point to the same ICA-reduced dataset if the requirement of taking absolute error into account priority cannot be metWith MSE, R is applied if taking into account capture priority of data laws2GCV should be used if the main objective is to consider reducing the overfitting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911324065.0A CN110969370B (en) | 2020-01-14 | 2020-01-14 | Quality risk analysis method for building structural member |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911324065.0A CN110969370B (en) | 2020-01-14 | 2020-01-14 | Quality risk analysis method for building structural member |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110969370A true CN110969370A (en) | 2020-04-07 |
| CN110969370B CN110969370B (en) | 2023-05-02 |
Family
ID=70035412
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911324065.0A Active CN110969370B (en) | 2020-01-14 | 2020-01-14 | Quality risk analysis method for building structural member |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110969370B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113408927A (en) * | 2021-06-30 | 2021-09-17 | 四川交达预应力工程检测科技有限公司 | Big data-based prestressed construction quality evaluation method and system |
| CN113592090A (en) * | 2021-09-27 | 2021-11-02 | 中建八局第三建设有限公司 | Building quality prediction method and device based on deep learning and storage medium |
| CN113962407A (en) * | 2021-10-19 | 2022-01-21 | 河南工业大学 | Multidimensional feature analysis method for monitoring data of active civil buildings |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105517019A (en) * | 2015-12-17 | 2016-04-20 | 南京华苏科技股份有限公司 | Method for detecting LTE (Long Term Evolution) network performance by using integrated regression system |
| CN106407498A (en) * | 2016-08-03 | 2017-02-15 | 无锡金世纪国民体质与健康研究有限公司 | Falling risk model establishment method |
| CN107451101A (en) * | 2017-07-21 | 2017-12-08 | 江南大学 | It is a kind of to be layered integrated Gaussian process recurrence soft-measuring modeling method |
| CN110110981A (en) * | 2019-04-26 | 2019-08-09 | 重庆第二师范学院 | A kind of credit rating Default Probability estimates and method for prewarning risk |
| CN110147321A (en) * | 2019-04-19 | 2019-08-20 | 北京航空航天大学 | A kind of recognition methods of the defect high risk module based on software network |
-
2020
- 2020-01-14 CN CN201911324065.0A patent/CN110969370B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105517019A (en) * | 2015-12-17 | 2016-04-20 | 南京华苏科技股份有限公司 | Method for detecting LTE (Long Term Evolution) network performance by using integrated regression system |
| CN106407498A (en) * | 2016-08-03 | 2017-02-15 | 无锡金世纪国民体质与健康研究有限公司 | Falling risk model establishment method |
| CN107451101A (en) * | 2017-07-21 | 2017-12-08 | 江南大学 | It is a kind of to be layered integrated Gaussian process recurrence soft-measuring modeling method |
| CN110147321A (en) * | 2019-04-19 | 2019-08-20 | 北京航空航天大学 | A kind of recognition methods of the defect high risk module based on software network |
| CN110110981A (en) * | 2019-04-26 | 2019-08-09 | 重庆第二师范学院 | A kind of credit rating Default Probability estimates and method for prewarning risk |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113408927A (en) * | 2021-06-30 | 2021-09-17 | 四川交达预应力工程检测科技有限公司 | Big data-based prestressed construction quality evaluation method and system |
| CN113592090A (en) * | 2021-09-27 | 2021-11-02 | 中建八局第三建设有限公司 | Building quality prediction method and device based on deep learning and storage medium |
| CN113962407A (en) * | 2021-10-19 | 2022-01-21 | 河南工业大学 | Multidimensional feature analysis method for monitoring data of active civil buildings |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110969370B (en) | 2023-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ghosn et al. | Reliability-based performance indicators for structural members | |
| Cavaleri et al. | Cyclic response of masonry infilled RC frames: Experimental results and simplified modeling | |
| Altarejos-García et al. | Methodology for estimating the probability of failure by sliding in concrete gravity dams in the context of risk analysis | |
| CN110969370B (en) | Quality risk analysis method for building structural member | |
| CN107515025B (en) | Ancient city wall disease grade evaluation method | |
| CN110991918B (en) | Safety risk assessment method in existing building construction project | |
| Beconcini et al. | Probabilistic reliability assessment of a heritage structure under horizontal loads | |
| Liu et al. | Statistical calibration of federal highway administration simplified models for facing tensile forces of soil nail walls | |
| Mata et al. | Correlation between compressive strength of concrete and ultrasonic pulse velocity: A case of study and a new correlation method | |
| Krentowski et al. | Historical masonry buildings’ condition assessment by non-destructive and destructive testing | |
| Orzhekhovskiy et al. | Design of stadium roofs with a given level of reliability | |
| Huang et al. | Multi-hazard coupling vulnerability analysis for buckling failure of vertical storage tank: Floods and hurricanes | |
| Ferlito et al. | Assessment of a vulnerability model against post-earthquake damage data: The case study of the historic city centre of L’Aquila in Italy | |
| CN115544794B (en) | Support steel frame design parameter selection method based on fatigue life prediction | |
| Shekhar et al. | Influence of corrosion on failure modes and lifetime seismic vulnerability assessment of low‐ductility RC frames | |
| CN115795599A (en) | Civil defense engineering structure earthquake vulnerability assessment method | |
| Rathje et al. | A seismic fragility framework for earth dams | |
| Opabola et al. | Repairability-based design of concrete structures | |
| Yang et al. | Time-dependent seismic reliability analysis of rocking cold-formed steel frame based on improved HOMM | |
| Basu et al. | Fragility analysis of drilled shafts in sand–an analytical correspondence between reliability index and factor of safety | |
| Farahani et al. | Probabilistic corrosion growth models of buried steel pipelines using inspection and soil survey data | |
| Gusev et al. | Research on regularities of deformation behavior of building structures in the areas of technogenic impact caused by mining | |
| Varela et al. | Evaluation of the structural reliability of a fixed offshore structure over time | |
| Shao et al. | Estimating potential economic loss from damage to RC buildings in Singapore due to far-field seismic risk | |
| Gu et al. | Performance assessment of a concrete gravity dam at Shenwo reservoir of China using deterministic and probabilistic methods |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |