CN104766129A - Subway shield construction surface deformation warning method based on temporal and spatial information fusion - Google Patents
Subway shield construction surface deformation warning method based on temporal and spatial information fusion Download PDFInfo
- Publication number
- CN104766129A CN104766129A CN201410848012.XA CN201410848012A CN104766129A CN 104766129 A CN104766129 A CN 104766129A CN 201410848012 A CN201410848012 A CN 201410848012A CN 104766129 A CN104766129 A CN 104766129A
- Authority
- CN
- China
- Prior art keywords
- temporal
- spatial
- alert
- earth
- shield
- 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.)
- Pending
Links
Abstract
The invention discloses a subway shield construction surface deformation warning method based on temporal and spatial information fusion. According to the method, shield construction surface deformation warning sign fusion and regional warning decision fusion are realized by combining an information fusion theory and a hierarchical fusion thought with surface deformation monitoring information and through a feature fusion algorithm based on a RBF neural network and a decision fusion algorithm based on a D-S evidence theory. A warning means and informationization tool integrating the functions of alarm identification, alarm analysis, alarm prediction, alarm evaluation and alarm decision is provided for the construction site. The method is of important significance to the enhancement of shield construction surface deformation safety risk monitoring and the improvement of shield construction safety risk management level.
Description
Technical field
The invention belongs to the early warning technology field of shield-tunneling construction earth's surface distortion, relate to the method for early warning that a kind of subway shield tunnel construction earth's surface based on Spatial-temporal Information Fusion is out of shape particularly.
Background technology
At present, utilize both at home and abroad about early warning technology comprehensive exploitation regarding safety early warning system and the ripe instrument applied in actual shield tunnel project development rapidly.GeoDATA company of Italy is proposed the information system management platform of GDMS by name (geodata master system), this system has used GIS and WEB technology, by 5 sub-System's composition such as buildings system for managing state, architectural risks evaluating system, shield structure data management system, Monitoring Data Management System and document file management systems, be applied in engineering is awaited a subway in St. Petersburg, Russia, Rome, ITA and Santiago respectively.TaiWan, China sub-newly built construction consultant incorporated company develops IDEAL monitoring materials disposal system for Taiwan underground engineering construction safety problem, this system is considered more comprehensive to Monitoring Data validity check and process, but the visualization function of monitoring information is more weak.The scholars such as Zhu Hehua propose to adopt and set up shield tunnel digital Platform based on client-server mode, give shield tunnel Data classification system and shield tunnel digital model, describe the modeling method of shield tunnel and surrounding formation thereof, and give visual and that information management is concrete implementation.For shield method tunnel construction, Shanghai Tunnel Engineering stock Co., Ltd develops shield method tunnel construction intelligent management system, under the prerequisite grasping construction information, pass through data analysis, engineering construction is effectively managed and technical support, and achieve real achievement in large-scale shield tunnel construction project.Shanghai is both " the peace journey subway engineering remote control administrative system " of the exploitation of engineering science and technology company limited, the technology such as transmission Network Based, wireless telecommunications, network data base, data analysis and automatic Prediction early warning, combine the much informations such as construction, management, monitoring, management and multimedia, be applied in Shanghai Underground engineering.
The method for early warning of subway shield tunnel construction earth's surface distortion both at home and abroad carries out early warning for single or individual event Monitoring Data mostly at present, lacks the fusion of space time information, causes the one-sidedness that early warning judges, affect the accuracy of early warning; Present stage, the theoretical foundation that obtains of predicted data was traditional stochastic medium theory, but directly used stochastic medium theory to carry out calculating to there is systematic error; Prewarning area division, on-the-spot inspection method etc. are Criterion or method not, the rationality of range of influence early warning and accuracy.
Therefore, invent a kind of science, effectively based on the method for early warning that the subway shield tunnel construction earth's surface of Spatial-temporal Information Fusion is out of shape, there is important engineering significance and realistic price.
Summary of the invention
The present invention is directed to the deficiencies in the prior art, the method for early warning that a kind of subway shield tunnel construction earth's surface based on Spatial-temporal Information Fusion is out of shape is provided.The method is according to Theory of Information Fusion, construct the shield-tunneling construction earth's surface distortion Early-warning Model based on Spatial-temporal Information Fusion, in conjunction with earth's surface deformation monitor information, by the Feature Fusion Algorithm based on RBF neural and the Decision fusion algorithm based on D-S evidence theory, achieve alert million fusions of shield-tunneling construction earth's surface distortion, regional early warning Decision fusion respectively.For working-yard provides, there is alert identification, alert analysis, alert prediction, alert evaluation, alert decision-making and early warning means integrally and information tool, for reinforcement shield-tunneling construction earth's surface transformation safe risk monitoring and control, improve shield-tunneling construction safety risk management level significant.
The invention provides the method for early warning that a kind of subway shield tunnel construction earth's surface based on Spatial-temporal Information Fusion is out of shape to comprise the following steps:
(1) prewarning area divides
The rule of construction area in work point range according to distortion temporal-spatial evolution is rationally decomposed, using engineering, environmental entity and the work progress after decomposition as the unit of alert appraisal and decision-making, dividing mode is as follows: microdeformation region, and the front 1 times of edpth of tunnel of shield structure face is with forefoot area; Distortion sharp increase region, 3 times of edpth of tunnel regions after shield structure face front 1 times of edpth of tunnel to shield structure face; Be out of shape slow region, 5 times of edpth of tunnel regions after 3 times of edpth of tunnels to shield structure face after shield structure face; Stabilization region, after shield structure face, 5 times of edpth of tunnels are with rear region;
(2) data acquisition
Comprise the acquisition of measured data, predicted data and tour data three class data, wherein measured data refers to the Monitoring Data of geologic parameter, construction parameter, design parameter, engineering environment information, each monitoring point; Predicted data refers to, based on construction earth's surface distortion temporal-spatial evolution Intelligent Forecasting, utilize the measured data monitored of monitoring point, realizes shield-tunneling construction earth's surface distortion temporal-spatial evolution intelligent predicting, obtains the predicted data of ground settlement, cross section deformation space distribution; Make an inspection tour the data message that data comprise on-the-spot tour project, safety monitoring project;
(3) alert sign merges
For each alert sign subspace, alert million subspaces of lower abbreviation are designed a police million and are merged RBF neural, complete the Nonlinear Mapping relation of respective police million subspace to alert state space respectively;
(4) early warning decision merges
Utilize the Output rusults of each RBF neural as the basic trust partition function of evidence theory, use D-S evidence theory compositional rule to merge these alert state spaces, obtain the regional early warning result of decision.
In technique scheme, the construction earth's surface distortion temporal-spatial evolution Intelligent Forecasting described in step (2), comprises the following steps:
Step one, measured data is collected, and comprises each monitoring point ground settlement, cross section deformation spatial distribution data, for systematic learning and test sample;
Step 2, builds temporal-spatial evolution computation model, builds the calculated relationship between earth's surface distortion temporal-spatial evolution characteristics parameter and surface deformation monitoring data; Based on improving the spatial distribution model of stochastic medium theory, calculating the final steady-state deformation value of shield-tunneling construction earth's surface any point, and being substituted in time course model, set up shield tunnel earth's surface any point (
x,
y,
z) distortion with the temporal-spatial evolution computation model of shield driving; Described earth's surface distortion temporal-spatial evolution characteristics parameter, comprises time course parameter and spatial distributed parameters two class; Time course parameter is made up of the time series parameters in scale parameter, form parameter, ARIMA; The Stratum Loss two kinds that spatial distributed parameters then comprises the major effect angle of overburden layer, shield-tunneling construction causes;
Step 3, builds temporal-spatial evolution relational model, based on the RBF neural algorithm of multiple TSP question population, constructs the temporal-spatial evolution relation between the input variable of earth's surface distortion temporal-spatial evolution system and shield structure earth's surface distortion temporal-spatial evolution characteristics parameter;
Step 4, surface deformation monitoring data prediction, utilizes measured data, temporal-spatial evolution computation model and temporal-spatial evolution relational model to carry out prognoses system study and test sample, predicts the data of ground settlement, cross section deformation space distribution etc.
In technique scheme, the acquisition methods making an inspection tour data in step (2) comprises the following steps:
Step one, divide make an inspection tour region, refer to engineering site make an inspection tour for region, the criteria for classifying is with the prewarning area criteria for classifying;
Step 2, makes an inspection tour content and comprises floor synthetic situation, surrounding building situation and underground utilities situation three class;
Step 3, makes an inspection tour result and judges, be divided into normal and abnormal two classes, represents respectively by numeral 0 and 1.
In technique scheme, the alert sign fusion method concrete operation step described in step (3) is as follows:
Step one, determines alert sign subspace, comprises surface deformation monitoring dotted state, the prediction of earth's surface distortion temporal-spatial evolution, shield-tunneling construction on-the-spot risk tour project status three sub spaces;
Step 2, determines alert state space, comprises ground fractures alert, surface collapse alert, construction cracking alert, building inclination alert and pipeline seepage alert five dimensions;
Step 3, by accumulating the learning training of a large amount of engineering sample, determines the RBF neural structure and parameter that each police million subspace is merged;
Step 4, utilizes RBF neural to build the Nonlinear Mapping relation of above-mentioned three police million subspaces to alert state space respectively, realizes alert million and merges.
In technique scheme, the early warning decision fusion method concrete operation step described in step (4) is as follows:
Step one, exports alert state space result by each police million subspace by RBF neural;
Step 2, utilizes the Output rusults of RBF neural as the basic trust partition function of evidence theory;
Step 3, uses D-S evidence theory compositional rule to merge these alert state spaces, obtains the regional early warning result of decision.
The method for early warning that the subway shield tunnel construction earth's surface based on Spatial-temporal Information Fusion that the present invention proposes is out of shape, overcome the problem of present stage early warning object simplification, achieve the fusion of space time information, according to Theory of Information Fusion and hierarchical fusion thinking, construct the shield-tunneling construction earth's surface distortion Early-warning Model based on Spatial-temporal Information Fusion, in conjunction with earth's surface deformation monitor information, by the Feature Fusion Algorithm based on RBF neural and the Decision fusion algorithm based on D-S evidence theory, achieve alert million fusions of shield-tunneling construction earth's surface distortion respectively, regional early warning Decision fusion, the theoretical foundation that whole model has science complete, for strengthening shield-tunneling construction earth's surface transformation safe risk monitoring and control, improve shield-tunneling construction safety risk management level significant.
Accompanying drawing explanation
Fig. 1 is the Early-warning Model schematic diagram of the inventive method.
Embodiment
Below in conjunction with drawings and the specific embodiments, the inventive method is further described.
As shown in Figure 1, the method for early warning that the present embodiment provides a kind of subway shield tunnel construction earth's surface based on Spatial-temporal Information Fusion to be out of shape comprises the following steps:
(1) prewarning area divides
The rule of construction area in work point range according to distortion temporal-spatial evolution is rationally decomposed, using engineering, environmental entity and the work progress after decomposition as the unit of alert appraisal and decision-making, dividing mode is as follows: microdeformation region, and the front 1 times of edpth of tunnel of shield structure face is with forefoot area; Distortion sharp increase region, 3 times of edpth of tunnel regions after shield structure face front 1 times of edpth of tunnel to shield structure face; Be out of shape slow region, 5 times of edpth of tunnel regions after 3 times of edpth of tunnels to shield structure face after shield structure face; Stabilization region, after shield structure face, 5 times of edpth of tunnels are with rear region;
(2) data acquisition
Comprise the acquisition of measured data, predicted data and tour data three class data, wherein measured data refers to the Monitoring Data of geologic parameter, construction parameter, design parameter, engineering environment information, each monitoring point; Predicted data refers to, based on construction earth's surface distortion temporal-spatial evolution Intelligent Forecasting, utilize the measured data monitored of monitoring point, realizes shield-tunneling construction earth's surface distortion temporal-spatial evolution intelligent predicting, obtains the predicted data of ground settlement, cross section deformation space distribution; Make an inspection tour the data message that data comprise on-the-spot tour project, safety monitoring project;
(3) alert sign merges
Design a police million for each alert million subspace and merge RBF neural, complete the Nonlinear Mapping relation of respective police million subspace to alert state space respectively;
(4) early warning decision merges
Utilize the Output rusults of each RBF neural as the basic trust partition function of evidence theory, use D-S evidence theory compositional rule to merge these alert state spaces, obtain the regional early warning result of decision.
In technique scheme, the construction earth's surface distortion temporal-spatial evolution Intelligent Forecasting described in step (2), comprises the following steps:
Step one, measured data is collected, and comprises each monitoring point ground settlement, cross section deformation spatial distribution data, for systematic learning and test sample;
Step 2, builds temporal-spatial evolution computation model, constructs the calculated relationship between earth's surface distortion temporal-spatial evolution characteristics parameter and surface deformation monitoring data; Based on improving the spatial distribution model of stochastic medium theory, calculating the final steady-state deformation value of shield-tunneling construction earth's surface any point, and being substituted in time course model, set up shield tunnel earth's surface any point (
x,
y,
z) distortion with the temporal-spatial evolution computation model of shield driving; Described earth's surface distortion temporal-spatial evolution characteristics parameter, comprises time course parameter and spatial distributed parameters two class; Time course parameter is made up of the time series parameters in scale parameter, form parameter, ARIMA; The Stratum Loss two kinds that spatial distributed parameters then comprises the major effect angle of overburden layer, shield-tunneling construction causes;
Step 3, builds temporal-spatial evolution relational model, based on the RBF neural algorithm of multiple TSP question population, constructs the temporal-spatial evolution relation between the input variable of earth's surface distortion temporal-spatial evolution system and shield structure earth's surface distortion temporal-spatial evolution characteristics parameter;
Step 4, surface deformation monitoring data prediction, utilizes measured data, temporal-spatial evolution computation model and temporal-spatial evolution relational model to carry out prognoses system study and test sample, predicts the data of ground settlement, cross section deformation space distribution etc.
In technique scheme, described time course model construction step is as follows:
Step 1: utilize cubic spline functions to carry out pre-service to shield-tunneling construction earth's surface deformation time series, obtains equidistantly, meets the shield-tunneling construction earth's surface deformation time series of analyzing samples amount requirement.
Step 2: the modeling of Weibull growth curve is carried out to pretreated shield-tunneling construction earth's surface deformation time series, use Non-linear parameter estimation method Confirming model parameter, its Weibull Fitting curve equation is:
(1)
Wherein,
for shield structure earth's surface distortion end value,
for shield structure earth's surface distortion initial value,
for shield structure earth's surface deformation curve scale parameter,
for shield structure earth's surface deformation curve form parameter.
Step 3: carry out stationary test to the residual sequence after Weibull Curve of growth fitting, meets stationary test, turns to step 5.
Step 4: to non-stationary regression criterion sequence, select suitable exponent number
carry out calculus of differences, until sequence meets stationary test after difference.
Step 5: white noise verification is carried out to differentiated regression criterion sequence, meets white noise verification and then turn to step 10.
Step 6: to differentiated regression criterion sequence, according to the character of sample correlation Coefficient Function ACF and sample PARCOR coefficients function PACF, determines the auto-correlation exponent number of arma modeling
with moving average exponent number
.
Step 7: right
in model
individual unknown parameter is estimated.
Step 8: inspection
the validity of model, if model is not by inspection, turns to step 6, reselects model matching again.
Step 9: if
model, by inspection, turns to step 6, takes into full account various possibility, sets up multiple model of fit, from all by selecting optimization model the model of fit of inspection.
Step 10: the Weibull-ARIMA model utilizing matching, the future trend of prediction shield structure earth's surface deformation time series.
In technique scheme, described is as follows based on the spatial distribution model computing method improving stochastic medium theory:
For axial along shield tunnel
axle arbitrary coordinate
the transversal section at place, shield tunnel buried depth is
, tunnel excavation radius is
, construction major effect angle is
, the Stratum Loss parameter that work progress causes is
, when shield tunnel construction causes the soil deformation of this section to stablize, any point in this transversal section
vertical final distortion computing formula be:
(2)
In technique scheme, described temporal-spatial evolution computation model computing method are as follows:
Set up shield tunnel earth's surface any point
distortion with the temporal-spatial evolution computation model of shield driving:
(3)
In formula,
integrating range cotype (2);
for Weibull Curve of growth fitting residual error;
;
;
for
rank calculus of differences.
In technique scheme, the described RBF neural algorithm steps based on multiple TSP question population is as follows:
Step 1: the constrained input variable determining RBF neural, and to training sample set
carry out standardization.
Step 2: utilize clustering algorithm determination RBF neural hidden layer node number, structure RBF neural model.
Step 3: according to center matrix, the width vector of basis function in RBF neural; Hidden layer, to the number of parameters such as connection weight matrix, threshold vector of output layer, determines the search volume dimension of algorithm
, and using the position coordinates of parameter vector as particle
, produce scale at random
n pprimary colony.
Step 4: algorithm correlation parameter is set: maximum iteration time
k max, the adaptive value limits of error
ε, flying speed of partcles limit value
v maxwith position limit value
x max, each particle initial position
x i 0and initial velocity
v i 0deng.
Step 5: be at iterations
ktime, parameter vector corresponding for each particle is mapped formation RBF neural.Calculate the fitness value of each particle of current population.
Step 6: to each particle, if it is more excellent with the current adaptive value of the best adaptive value of its individual history to compare its current adaptive value, then upgrades the best adaptive value of its individual history and desired positions
pbest; Compare current adaptive value and the best adaptive value of global history of all particles in colony, if the current adaptive value of certain particulate is more excellent, then upgrade the best adaptive value of global history and desired positions
gbest.
Step 7: TSP question operation is performed to parameter and population.According to average fitness and the population global optimum fitness of formula (3), (4) calculating population, and calculate evolution equation parameter according to formula (5), (6), (7)
ω,
c 1,
c 2.
Step 8: order
k=
k+ 1, the speed of each particle is upgraded according to formula (6)
v i k+ 1
and position
x i k+ 1
, after amplitude limiting processing, namely adjust RBF neural model parameter.
din dimension search volume, for particle
i's
(
) dimension exist
kthe evolution equation in+1 moment is:
(9)
Step 9: to population
pbestexecution TSP question operates.Colony fitness variance is calculated according to formula (9) and (10)
δ k 2, and calculate global optimum position according to formula (11)
gbestmutation probability
p m .Produce stochastic variable
rand()
(0,1), when
rand() <
p m time perform global optimum position by formula (12)
gbestmutation operation; Otherwise turn to step 10.
Step 10: re-start fitness evaluation to the new particle population that iteration produces, (adaptive value error reaches to judge whether to meet stop condition
or iterations exceedes
k max), if met, algorithm terminates.Otherwise return Step5 and continue search.
In technique scheme, the acquisition methods making an inspection tour data in step (2) comprises the following steps:
Step one, divide make an inspection tour region, refer to engineering site make an inspection tour for region, the criteria for classifying is with the prewarning area criteria for classifying;
Step 2, makes an inspection tour content and comprises floor synthetic situation, surrounding building situation and underground utilities situation three class;
Step 3, makes an inspection tour result and judges, be divided into normal and abnormal two classes, represents respectively by numeral 0 and 1.
In technique scheme, the alert sign fusion method concrete operation step described in step (3) is as follows:
Step one, determines alert sign subspace, comprises surface deformation monitoring dotted state, the prediction of earth's surface distortion temporal-spatial evolution, shield-tunneling construction on-the-spot risk tour project status three sub spaces;
Step 2, determines alert state space, comprises ground fractures alert, surface collapse alert, construction cracking alert, building inclination alert and pipeline seepage alert five dimensions;
Step 3, by accumulating the learning training of a large amount of engineering sample, determines the RBF neural structure and parameter that each police million subspace is merged;
Step 4, utilizes RBF neural to build the Nonlinear Mapping relation of above-mentioned three police million subspaces to alert state space respectively, realizes alert million and merges.
In technique scheme, the early warning decision fusion method concrete operation step described in step (4) is as follows:
Step one, exports alert state space result by each police million subspace by RBF neural;
Step 2, utilizes the Output rusults of RBF neural as the basic trust partition function of evidence theory;
Step 3, uses D-S evidence theory compositional rule to merge these alert state spaces, obtains the regional early warning result of decision.
In technique scheme, the mathematical model of evidence theory is:
Step 1: first establish identification framework Θ.Only establish framework Θ the research for proposition could be converted into set research.
Step 2: according to the original allocation of an Evidence For Establishing reliability, namely evidence treatment people is analyzed evidence, determines the degree of support of evidence to each set (proposition) itself.The large I that basic trust distributes is provided by rule of thumb by people, or according to data configuration.
Step 3: analyze causes and effects, calculates the reliability for all propositions.
Because traditional evidence fusion method faces the problem such as " veto by one vote " and " robustness ", the Dempster-Shafer evidence theory therefore according to amendment model combines evidence:
Step 1: according to the conflict factor, judge whether evidence source has conflict, if Lothrus apterus, then directly carry out fusion treatment with Dempster rule of combination, otherwise carry out next step.Wherein, Dempster rule of combination is as follows:
（14）
Wherein,
m (A)be called that evidence is to proposition
abasic Probability As-signment or substantially credible number.
krepresent evidence
between conflicting information.
Step 2: based on BPA and the burnt meta-attribute thereof in evidence source, the distance function designed by Jousselme, calculates the weights of evidence, carries out pre-service to evidence.
Step 3: the weighted mean evidence calculating evidence source.
Step 4: average evidence is replaced the conflicting evidence in evidence source, and inherits corresponding weight value.
Step 5: evidence model is modified, and then combine according to Dempster rule of combination.
Step 6: the BPA of Jiao unit after combination is normalized.
Claims (5)
1., based on the method for early warning that the subway shield tunnel construction earth's surface of Spatial-temporal Information Fusion is out of shape, it is characterized in that comprising the following steps:
(1) prewarning area divides
The rule of construction area in work point range according to distortion temporal-spatial evolution is rationally decomposed, using engineering, environmental entity and the work progress after decomposition as the unit of alert appraisal and decision-making, dividing mode is as follows: microdeformation region, and the front 1 times of edpth of tunnel of shield structure face is with forefoot area; Distortion sharp increase region, 3 times of edpth of tunnel regions after shield structure face front 1 times of edpth of tunnel to shield structure face; Be out of shape slow region, 5 times of edpth of tunnel regions after 3 times of edpth of tunnels to shield structure face after shield structure face; Stabilization region, after shield structure face, 5 times of edpth of tunnels are with rear region;
(2) data acquisition
Comprise the acquisition of measured data, predicted data and tour data three class data, wherein measured data refers to the Monitoring Data of geologic parameter, construction parameter, design parameter, engineering environment information, each monitoring point; Predicted data refers to, based on construction earth's surface distortion temporal-spatial evolution Intelligent Forecasting, utilize the measured data monitored of monitoring point, realizes shield-tunneling construction earth's surface distortion temporal-spatial evolution intelligent predicting, obtains the predicted data of ground settlement, cross section deformation space distribution; Make an inspection tour the data message that data comprise on-the-spot tour project, safety monitoring project;
(3) alert sign merges
Design a police million for each alert sign subspace and merge RBF neural, complete the Nonlinear Mapping relation of respective police million subspace to alert state space respectively;
(4) early warning decision merges
Utilize the Output rusults of each RBF neural as the basic trust partition function of evidence theory, use D-S evidence theory compositional rule to merge these alert state spaces, obtain the regional early warning result of decision.
2. the method for early warning that is out of shape of a kind of subway shield tunnel construction earth's surface based on Spatial-temporal Information Fusion according to claim 1, is characterized in that the construction earth's surface distortion temporal-spatial evolution Intelligent Forecasting described in step (2), comprises the following steps:
Step one, measured data is collected, and comprises each monitoring point ground settlement, cross section deformation spatial distribution data, for systematic learning and test sample;
Step 2, builds temporal-spatial evolution computation model, constructs the calculated relationship between earth's surface distortion temporal-spatial evolution characteristics parameter and surface deformation monitoring data; Based on improving the spatial distribution model of stochastic medium theory, calculating the final steady-state deformation value of shield-tunneling construction earth's surface any point, and being substituted in time course model, set up shield tunnel earth's surface any point (
x,
y,
z) distortion with the temporal-spatial evolution computation model of shield driving; Described earth's surface distortion temporal-spatial evolution characteristics parameter, comprises time course parameter and spatial distributed parameters two class; Time course parameter is made up of the time series parameters in scale parameter, form parameter, ARIMA; The Stratum Loss two kinds that spatial distributed parameters then comprises the major effect angle of overburden layer, shield-tunneling construction causes;
Step 3, builds temporal-spatial evolution relational model, based on RBF neural algorithm, constructs the temporal-spatial evolution relation between the input variable of earth's surface distortion temporal-spatial evolution system and shield structure earth's surface distortion temporal-spatial evolution characteristics parameter;
Step 4, surface deformation monitoring data prediction, utilizes measured data, temporal-spatial evolution computation model and temporal-spatial evolution relational model to carry out prognoses system study and test sample, predicts the data of ground settlement, cross section deformation space distribution etc.
3. the method for early warning that is out of shape of a kind of subway shield tunnel construction earth's surface based on Spatial-temporal Information Fusion according to claim 1, is characterized in that the acquisition methods making an inspection tour data in step (2) comprises the following steps:
Step one, divide make an inspection tour region, refer to engineering site make an inspection tour for region, the criteria for classifying is with the prewarning area criteria for classifying;
Step 2, makes an inspection tour content and comprises floor synthetic situation, surrounding building situation and underground utilities situation three class;
Step 3, makes an inspection tour result and judges, be divided into normal and abnormal two classes, represents respectively by numeral 0 and 1.
4. the method for early warning that is out of shape of a kind of subway shield tunnel construction earth's surface based on Spatial-temporal Information Fusion according to claim 1, is characterized in that the alert sign fusion method concrete operation step described in step (3) is as follows:
Step one, determines alert sign subspace, comprises surface deformation monitoring dotted state, the prediction of earth's surface distortion temporal-spatial evolution, shield-tunneling construction on-the-spot risk tour project status three sub spaces;
Step 2, determines alert state space, comprises ground fractures alert, surface collapse alert, construction cracking alert, building inclination alert and pipeline seepage alert five dimensions;
Step 3, by accumulating the learning training of a large amount of engineering sample, determines the RBF neural structure and parameter that each police million subspace is merged;
Step 4, utilizes RBF neural to build the Nonlinear Mapping relation of above-mentioned three police million subspaces to alert state space respectively, realizes alert million and merges.
5. the method for early warning that is out of shape of a kind of subway shield tunnel construction earth's surface based on Spatial-temporal Information Fusion according to claim 1, is characterized in that the early warning decision fusion method concrete operation step described in step (4) is as follows:
Step one, exports alert state space result by each police million subspace by RBF neural;
Step 2, utilizes the Output rusults of RBF neural as the basic trust partition function of evidence theory;
Step 3, uses D-S evidence theory compositional rule to merge these alert state spaces, obtains the regional early warning result of decision.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410848012.XA CN104766129A (en) | 2014-12-31 | 2014-12-31 | Subway shield construction surface deformation warning method based on temporal and spatial information fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410848012.XA CN104766129A (en) | 2014-12-31 | 2014-12-31 | Subway shield construction surface deformation warning method based on temporal and spatial information fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104766129A true CN104766129A (en) | 2015-07-08 |
Family
ID=53647943
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410848012.XA Pending CN104766129A (en) | 2014-12-31 | 2014-12-31 | Subway shield construction surface deformation warning method based on temporal and spatial information fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104766129A (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021717A (en) * | 2016-05-19 | 2016-10-12 | 辽宁工程技术大学 | Neural network-based method for analyzing surface subsidence caused by metro excavation |
| CN106295057A (en) * | 2016-08-23 | 2017-01-04 | 中南大学 | A kind of earth pressure balanced shield, EPBS Tu Cang enters the Ground surface settlement method that casting causes |
| CN106779296A (en) * | 2016-11-22 | 2017-05-31 | 华中科技大学 | A kind of constructing tunnel Adjacent Buildings safe early warning method based on multisensor |
| CN106897813A (en) * | 2017-01-13 | 2017-06-27 | 同济大学 | A kind of large shield tunnel intelligent management system based on BIM and AI |
| CN106909715A (en) * | 2017-01-19 | 2017-06-30 | 中铁开发投资有限公司 | The girder construction Deformation Prediction control method of underground structure is worn under a kind of tunnel |
| CN107092934A (en) * | 2017-04-25 | 2017-08-25 | 淮阴师范学院 | A kind of large scale structure damnification recognition method based on three-level data fusion |
| CN107433687A (en) * | 2017-07-13 | 2017-12-05 | 重庆交通建设(集团)有限责任公司 | Supporting and corollary equipment optimization method after a kind of |
| CN108985340A (en) * | 2018-06-22 | 2018-12-11 | 西安电子科技大学 | Shield construction ground predicting subsidence method based on dual model fusion |
| CN109359412A (en) * | 2018-11-01 | 2019-02-19 | 山东大学 | The calculation method and system that prediction tunneling shield digging process deforms entirely |
| CN110136374A (en) * | 2019-05-22 | 2019-08-16 | 微云（武汉）科技有限公司 | A kind of generation method, device and the storage medium of building evacuation route |
| CN110159346A (en) * | 2019-05-13 | 2019-08-23 | 辽宁工程技术大学 | Class rectangle tunnel Displacement forecast method based on non-uniform convergence mode |
| CN110442979A (en) * | 2019-08-08 | 2019-11-12 | 山东大学 | The full Deformation Prediction method and system in the shield-tunneling construction tunnel based on BP neural network |
| CN111353678A (en) * | 2018-12-24 | 2020-06-30 | 杭州海康威视数字技术股份有限公司 | Model generation method, device and equipment |
| CN111598460A (en) * | 2020-05-18 | 2020-08-28 | 武汉轻工大学 | Method, device and equipment for monitoring heavy metal content in soil and storage medium |
| CN112818567A (en) * | 2021-03-01 | 2021-05-18 | 深圳市安泰数据监测科技有限公司 | Geotechnical engineering intelligent monitoring and early warning method and device based on probability theory |
| CN113204824A (en) * | 2021-05-21 | 2021-08-03 | 上海大学 | Multi-model fusion shield construction settlement prediction method and system |
| CN113239439A (en) * | 2021-05-21 | 2021-08-10 | 上海大学 | Shield construction ground surface settlement prediction system and method |
| CN113310528A (en) * | 2021-06-16 | 2021-08-27 | 深圳防灾减灾技术研究院 | Real-time tunnel structure health monitoring method based on multivariate sensing data |
| CN113362469A (en) * | 2021-08-09 | 2021-09-07 | 长沙市轨道交通三号线建设发展有限公司 | Shield tunnel construction early warning method integrating city building information and stratum structure |
| CN113362389A (en) * | 2021-06-08 | 2021-09-07 | 长安大学 | Shield tunnel deformation prediction method based on CNN-RNN coupling number-shape fusion |
| CN113724105A (en) * | 2021-09-08 | 2021-11-30 | 姚成龙 | Building construction site monitoring system and monitoring method thereof |
| CN115412923A (en) * | 2022-10-28 | 2022-11-29 | 河北省科学院应用数学研究所 | Multi-source sensor data credible fusion method, system, equipment and storage medium |
| CN115758038A (en) * | 2022-11-08 | 2023-03-07 | 青岛理工大学 | Stratum composite ratio calculation method, device, equipment and medium for shield construction |
-
2014
- 2014-12-31 CN CN201410848012.XA patent/CN104766129A/en active Pending
Non-Patent Citations (1)
| Title |
|---|
| 周诚: "地铁盾构施工地表变形时空演化规律与预警研究", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106021717A (en) * | 2016-05-19 | 2016-10-12 | 辽宁工程技术大学 | Neural network-based method for analyzing surface subsidence caused by metro excavation |
| CN106295057B (en) * | 2016-08-23 | 2019-04-30 | 中南大学 | A kind of earth pressure balanced shield, EPBS Tu Cang is into Ground surface settlement method caused by casting |
| CN106295057A (en) * | 2016-08-23 | 2017-01-04 | 中南大学 | A kind of earth pressure balanced shield, EPBS Tu Cang enters the Ground surface settlement method that casting causes |
| CN106779296A (en) * | 2016-11-22 | 2017-05-31 | 华中科技大学 | A kind of constructing tunnel Adjacent Buildings safe early warning method based on multisensor |
| CN106897813A (en) * | 2017-01-13 | 2017-06-27 | 同济大学 | A kind of large shield tunnel intelligent management system based on BIM and AI |
| CN106909715A (en) * | 2017-01-19 | 2017-06-30 | 中铁开发投资有限公司 | The girder construction Deformation Prediction control method of underground structure is worn under a kind of tunnel |
| CN107092934A (en) * | 2017-04-25 | 2017-08-25 | 淮阴师范学院 | A kind of large scale structure damnification recognition method based on three-level data fusion |
| CN107433687A (en) * | 2017-07-13 | 2017-12-05 | 重庆交通建设(集团)有限责任公司 | Supporting and corollary equipment optimization method after a kind of |
| CN108985340B (en) * | 2018-06-22 | 2021-06-01 | 西安电子科技大学 | Shield construction ground settlement prediction method based on dual-model fusion |
| CN108985340A (en) * | 2018-06-22 | 2018-12-11 | 西安电子科技大学 | Shield construction ground predicting subsidence method based on dual model fusion |
| CN109359412A (en) * | 2018-11-01 | 2019-02-19 | 山东大学 | The calculation method and system that prediction tunneling shield digging process deforms entirely |
| CN109359412B (en) * | 2018-11-01 | 2021-04-16 | 山东大学 | Calculation method and system for predicting total deformation of tunnel shield excavation process |
| CN111353678A (en) * | 2018-12-24 | 2020-06-30 | 杭州海康威视数字技术股份有限公司 | Model generation method, device and equipment |
| CN110159346A (en) * | 2019-05-13 | 2019-08-23 | 辽宁工程技术大学 | Class rectangle tunnel Displacement forecast method based on non-uniform convergence mode |
| CN110159346B (en) * | 2019-05-13 | 2020-07-31 | 辽宁工程技术大学 | Quasi-rectangular tunnel earth surface deformation prediction method based on non-uniform convergence mode |
| CN110136374A (en) * | 2019-05-22 | 2019-08-16 | 微云（武汉）科技有限公司 | A kind of generation method, device and the storage medium of building evacuation route |
| CN110442979B (en) * | 2019-08-08 | 2021-04-13 | 山东大学 | BP neural network-based shield construction tunnel total deformation prediction method and system |
| CN110442979A (en) * | 2019-08-08 | 2019-11-12 | 山东大学 | The full Deformation Prediction method and system in the shield-tunneling construction tunnel based on BP neural network |
| CN111598460A (en) * | 2020-05-18 | 2020-08-28 | 武汉轻工大学 | Method, device and equipment for monitoring heavy metal content in soil and storage medium |
| CN111598460B (en) * | 2020-05-18 | 2023-09-29 | 武汉轻工大学 | Method, device, equipment and storage medium for monitoring heavy metal content of soil |
| CN112818567A (en) * | 2021-03-01 | 2021-05-18 | 深圳市安泰数据监测科技有限公司 | Geotechnical engineering intelligent monitoring and early warning method and device based on probability theory |
| CN113204824B (en) * | 2021-05-21 | 2023-04-07 | 上海大学 | Multi-model fusion shield construction settlement prediction method and system |
| CN113204824A (en) * | 2021-05-21 | 2021-08-03 | 上海大学 | Multi-model fusion shield construction settlement prediction method and system |
| CN113239439A (en) * | 2021-05-21 | 2021-08-10 | 上海大学 | Shield construction ground surface settlement prediction system and method |
| CN113239439B (en) * | 2021-05-21 | 2022-04-05 | 上海大学 | Shield construction ground surface settlement prediction system and method |
| CN113362389B (en) * | 2021-06-08 | 2024-02-27 | 长安大学 | Shield tunnel deformation prediction method based on CNN-RNN coupling number-shape fusion |
| CN113362389A (en) * | 2021-06-08 | 2021-09-07 | 长安大学 | Shield tunnel deformation prediction method based on CNN-RNN coupling number-shape fusion |
| CN113310528A (en) * | 2021-06-16 | 2021-08-27 | 深圳防灾减灾技术研究院 | Real-time tunnel structure health monitoring method based on multivariate sensing data |
| CN113362469B (en) * | 2021-08-09 | 2021-10-08 | 长沙市轨道交通三号线建设发展有限公司 | Shield tunnel construction early warning method integrating city building information and stratum structure |
| CN113362469A (en) * | 2021-08-09 | 2021-09-07 | 长沙市轨道交通三号线建设发展有限公司 | Shield tunnel construction early warning method integrating city building information and stratum structure |
| CN113724105A (en) * | 2021-09-08 | 2021-11-30 | 姚成龙 | Building construction site monitoring system and monitoring method thereof |
| CN115412923A (en) * | 2022-10-28 | 2022-11-29 | 河北省科学院应用数学研究所 | Multi-source sensor data credible fusion method, system, equipment and storage medium |
| CN115758038A (en) * | 2022-11-08 | 2023-03-07 | 青岛理工大学 | Stratum composite ratio calculation method, device, equipment and medium for shield construction |
| CN115758038B (en) * | 2022-11-08 | 2023-08-08 | 青岛理工大学 | Stratum composite ratio calculation method, device, equipment and medium for shield construction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104766129A (en) | Subway shield construction surface deformation warning method based on temporal and spatial information fusion | |
| Wu et al. | Depth prediction of urban flood under different rainfall return periods based on deep learning and data warehouse | |
| Pan et al. | Multi-classifier information fusion in risk analysis | |
| Wu et al. | Rock burst prediction probability model based on case analysis | |
| Lin et al. | Risk assessment and management of excavation system based on fuzzy set theory and machine learning methods | |
| Ghosh et al. | Urban ecological security assessment and forecasting using integrated DEMATEL-ANP and CA-Markov models: A case study on Kolkata Metropolitan Area, India | |
| Zhang et al. | An improved Dempster–Shafer approach to construction safety risk perception | |
| Wu et al. | A dynamic Bayesian network based approach to safety decision support in tunnel construction | |
| CN106407493B (en) | A kind of rock burst grade evaluation method based on multidimensional Gauss cloud model | |
| Leu et al. | Probabilistic prediction of tunnel geology using a Hybrid Neural-HMM | |
| CN102194056B (en) | BN-GIS (Bayesian Network-Geographic Information System) method for evaluating and predicting water inrush danger of coal-seam roof and floor | |
| Zhang et al. | Risk-based estimate for operational safety in complex projects under uncertainty | |
| Zhou et al. | Impacts of building configurations on urban stormwater management at a block scale using XGBoost | |
| Wang et al. | A risk evaluation method with an improved scale for tunnel engineering | |
| Leng et al. | A hybrid data mining method for tunnel engineering based on real-time monitoring data from tunnel boring machines | |
| Li et al. | International roughness index prediction based on multigranularity fuzzy time series and particle swarm optimization | |
| Gu et al. | Health and safety situation awareness model and emergency management based on multi-sensor signal fusion | |
| Wu et al. | Dynamic risk analysis for adjacent buildings in tunneling environments: a Bayesian network based approach | |
| Nedic et al. | Data mining with various optimization methods | |
| Lin et al. | A quantitative method to assess the ecological indicator system's effectiveness: a case study of the ecological province construction indicators of China | |
| Zhou et al. | Topological mapping and assessment of multiple settlement time series in deep excavation: a complex network perspective | |
| Dong et al. | Data analysis in pavement engineering: An overview | |
| Lin et al. | Ensemble model for risk status evaluation of excavation | |
| Wu et al. | A dynamic decision approach for risk analysis in complex projects | |
| Mahini et al. | Dynamic spatial modeling of urban growth through cellular automata in a GIS environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150708 |