CN107315862A - A kind of method for setting up open-cut foundation ditch engineering investigation and analog parameter relation - Google Patents
A kind of method for setting up open-cut foundation ditch engineering investigation and analog parameter relation Download PDFInfo
- Publication number
- CN107315862A CN107315862A CN201710438743.0A CN201710438743A CN107315862A CN 107315862 A CN107315862 A CN 107315862A CN 201710438743 A CN201710438743 A CN 201710438743A CN 107315862 A CN107315862 A CN 107315862A
- Authority
- CN
- China
- Prior art keywords
- parameter
- calculation
- soil
- neural network
- result
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Investigation Of Foundation Soil And Reinforcement Of Foundation Soil By Compacting Or Drainage (AREA)
Abstract
The present invention discloses a kind of method for setting up open-cut foundation ditch engineering investigation and analog parameter relation, gathers the Monitoring Data of base pit engineering；Set up suitable soil constitutive model；The original prospecting parameter of soil layer is extracted, and then expands to the sample parameter and test sample parameter of numerical computations；Sample parameter and test sample parameter are subjected to numerical computations respectively；The BP neural network that result of calculation is substituted into genetic algorithm optimization is trained, the mapping relations set up between Soil Parameters and numerical result, and tests neural network model with test sample parameter；The relation that actual monitoring data are substituted into Soil Parameters and result of calculation tries to achieve the corresponding Soil Parameters of measured displacements；Multiple base pit engineering cases are respectively calculated, the Soil Parameters of the prospecting parameter of multiple base pit engineering cases simulation corresponding with the measured displacements that inverting is obtained are substituted into the BP neural network of genetic algorithm optimization, the relation of open-cut foundation ditch engineering investigation and analog parameter is set up.
Description
Technical field
Open-cut foundation ditch engineering investigation and analog parameter are set up the invention belongs to technical field of civil engineering, more particularly to one kind
The method of relation, the analog parameter for obtaining closing to reality operating mode during open-cut foundation ditch engineering excavation by prospecting parameter, enters
And more accurately numerical computations are carried out to Practical Project.
Background technology
With the development of modern science and technology, Computer Numerical Simulation is playing increasing work in all trades and professions
With.In Geotechnical Engineering field, with large geotechnical structure, high-rise building and underground mining, tunnel over strait, city rail
A large amount of constructions of the underground engineerings such as traffic etc. so that application of the numerical simulation technology in Geotechnical Engineering is continued to develop, rock
Native engineering numerical is played a greater and greater role during project surveying, design, construction, and base pit engineering numerical simulation
It is its important part.
The credibility of numerical simulation calculation result and engineering design, the security of arrangement and method for construction depend primarily on numerical value
Calculate the degree of reliability with Rock And Soil parameter in simulation process.Because Rock And Soil is in forming process, mineralogical composition, soil layer are deep
The factors such as degree, stress history, water content and density are ever-changing, and the property of Rock And Soil may have bigger difference at each point.At present
It is typically with statistical analysis and with reference to empirical data from test data in the rock ＆ soil mechanical parameter needed for geotechnical engineering design
It is fixed, with certain arbitrariness.There is random uncertain feature as the test data of sampling, excursion is larger.When
When prospecting operation can not provide correct, rational Rock And Soil parameter, uncertainty will be brought to engineering design, construction safety,
It is easily caused the generation of engineering accident.
On the 10th national rock-soil mechanics numerical value and analytic method discussion, the one of Zhejiang University professor Gong Xiaonan initiation
Investigation on Geotechnical Engineering numerical analysis, respondent be distributed throughout the country be engaged in geotechnical study and engineering
All kinds of personnel of application, with certain authority.Professor Gong Xiaonan is classified statistics to investigation result.By investigation result
As can be seen that the research of the Geotechnical Engineering of participation investigation and engineers and technicians generally believe numerical simulation technology in Geotechnical Engineering
In occupy increasingly consequence, ratio is more than 90%.At the same time, when numerical analysis is used for geotechnical engineering problems, parameter
Determine and obtain high attention with constitutive model, be to influence the key factor of Numerical results reliability, numerical analysis side
Method includes numerical algorithm, constitutive relation research and selection, parametric measurement, initial and the determination of boundary condition and analysis skill etc.
Aspects of contents, in the research and engineers and technicians for participating in investigation, parametric measurement is 114 person-times, occupies 43.02% most
At high proportion.In view of the precision of current value analysis is also weak, for how further to improve ground numerical analysis ability needs
The investigation of the key issue of solution shows have 112 person-times to think the problem of Parameters of constitutive model measure is most critical, occupy
49.56%, far above other factors, it can be seen that parameter determines the importance in Geotechnical Engineering analysis.
Due to the complexity of rock soil medium property, directly by reconnoitre and test the parameter obtained usually contain it is many artificial and
Other error components so that be directly used in calculating analysis and be difficult to obtain and the preferable result of actual coincidence.Under this background, instead
Analysis method is arisen at the historic moment.Back analysis is i.e. with the data of the field monitoring such as information such as displacement, strain, stress or load, by building
Vertical inverse model, inverse obtains some initial parameters of the system, such as stratum primary stress, the method for Parameters of constitutive model.
The purpose of back analysis is to set up a theoretical prediction model closer to field actual measurement results, more correctly to reflect or in advance
Survey some mechanical behaviors of geotechnical structure.
At present, with the continuous progress of technological meanses, prospecting and experimental technique and back analysis algorithm obtain certain journey
The development of degree, reconnoitres precision getparms and improves constantly, back analysis numerical algorithm is also proposed and improved in succession.However, logical
The research for crossing relation between both the above means acquisition prospecting and analog parameter is but usually neglected, and is closed between both parameters
The foundation of system can improve the value ability of base pit engineering geotechnical engineering problems parameters for numerical simulation；Reduce Geotechnical Engineering modeling point
Analyse parameter and choose complexity, shorten and adjust the ginseng time, improve numerical simulation calculation precision and reliability, be more effectively design of foundation pit
There is provided and assess, check；Issuable Geotechnical Problems of constructing are predicted, assessed and early warning, carried by construction simulation process in advance
Before adopt an effective measure, probability etc. that effectively reduction (subway) accident occurs.
The content of the invention
Present invention aims at a kind of utilization numerical computations, the BP neural network of genetic algorithm optimization is provided, by actually supervising
The calculating parameter that result data inverting obtains the numerical model of closing to reality operating mode is surveyed, genetic algorithm optimization is then reused
The method that BP neural network sets up relation between prospecting parameter and analog parameter, closing to reality is can obtain by the prospecting parameter of engineering
The analog parameter of operating mode, and then obtain the numerical result with being actually consistent.
To achieve the above object, the present invention is adopted the following technical scheme that：
It is a kind of to set up open-cut foundation ditch engineering investigation and the method for analog parameter relation comprises the following steps：
Step 1, the Monitoring Data for gathering base pit engineering, are fitted to it according to its changing rule, are used as Inversion Calculation
Basic data；
Step 2, the mathematical calculation model for setting up standard, choose suitable soil constitutive model；The original of soil layer is extracted to survey
Parameter is examined, the initial soil calculation parameter of numerical computations is converted to this；Based on initial soil calculation parameter, number is expanded to
It is worth the sample parameter calculated and test sample parameter；Sample parameter and test sample parameter are subjected to numerical computations respectively；
Step 3, the BP neural network that sample parameter and numerical result are substituted into genetic algorithm optimization are trained, and are built
Vertical mapping relations between Soil Parameters and numerical result, and test neural network model with test sample parameter；Will
Actual monitoring data substitute into Soil Parameters and the relation of result of calculation tries to achieve the corresponding Soil Parameters of measured displacements；To multiple foundation ditches
Engineering Projects is respectively calculated, by the prospecting parameter of multiple base pit engineering cases mould corresponding with the measured displacements that inverting is obtained
The Soil Parameters of plan substitute into the BP neural network of genetic algorithm optimization, set up the pass of open-cut foundation ditch engineering investigation and analog parameter
System.
Preferably, the Monitoring Data in step 1 includes：Pile head settlement, ground settlement, Horizontal Displacement, pile body water
Prosposition is moved.
Preferably, the original prospecting parameter in step 1 includes：Cohesive strength, internal friction angle, modulus of compressibility, Poisson's ratio, day
Right density, natural moisture content, proportion, void ratio, liquid limit, plasticity index, cake compressibility, swelling index.
Preferably, step 3 specifically includes following steps：
Step 3.1：The result of calculation that step 2 is obtained substitutes into neural network model, by soil body sample parameter (Nai…Nfi)
It is used as input layer, its corresponding simulation result of calculation (Sai…Sfi) as output layer, node in hidden layer can useWherein, n
It is respectively to input and output layer nodes with m, using 2n+1 initial the number of hidden nodes, Population in Genetic Algorithms scale is set, changed
Generation number, intersection and mutation probability, then run neural network model, set up between soil body sample parameter and simulation result of calculation
Mapping relations；
Step 3.2：By test sample parameter (nai…nfi) substitute into soil body sample parameter and simulation meter that step 3.1 is set up
The mapping relations between result are calculated, the neural network model that operating procedure 3.1 is preserved is calculated, and show that test sample is corresponding
Neural computing result (Lai…Lfi), comparative analysis test sample parameter (nai…nfi) corresponding numerical simulation calculation result
(sai…sfi) and neural computing result (Lai…Lfi) error, meet require, then carry out next step；
Step 3.3：By actual monitoring data (Sa1…Sf1) substitute into step 3.1 set up neural network model, run and carry out
Calculate, and then draw by the Soil Parameters (M of the corresponding simulation of measured dataa1…Mf1), each open-cut foundation ditch engineering is according to upper
Flow is stated to be respectively calculated.The prospecting parameter of all base pit engineering cases is (ai…fi), by measured data according to above-mentioned steps
Soil Parameters (the M for the simulation that Inversion Calculation is obtainedai…Mfi)；
Step 3.4：Will prospecting parameter (ai…fi) as the input layer of neutral net, obtained by measured data Inversion Calculation
Simulation Soil Parameters (Mai…Mfi) as the output layer of neutral net, set neutral net and heredity to calculate according to step 3.1
The correlation computations parameter of method, operation neural network model is calculated, and then draws open-cut foundation ditch engineering investigation and analog parameter
Between mapping relation.
Beneficial effect
It is of the invention effective using numerical computations, the BP neural network of genetic algorithm optimization and actual monitoring, survey data,
The relation set up between prospecting and analog parameter, the value energy of base pit engineering geotechnical engineering problems parameters for numerical simulation is improved with this
Power；Reduce Geotechnical Engineering modeling analysis parameter and choose complexity, shorten and adjust the ginseng time, improve numerical simulation calculation precision and reliable
Property, more effectively provide assessment, check for design of foundation pit；Construction simulation process, in advance to issuable Geotechnical Problems of constructing
It is predicted, assesses and early warning, adopts an effective measure in advance, probability that effectively reduction (subway) accident occurs etc..
Brief description of the drawings：
The various constitutive models of Fig. 1 excavate the schematic diagram of applicability in analysis in foundation ditch numerical value；
Fig. 2 genetic algorithm optimization BP neural network flow charts；
Fig. 3 the inventive method flow charts.
Specific implementation method：
As shown in figure 3, the present invention provides a kind of method for setting up open-cut foundation ditch engineering investigation and analog parameter relation, including
Following steps：
1) selection of Engineering Projects, its control condition includes：
1.1) Engineering Projects needed for must include the data such as prospecting, design, third party monitoring and data is complete
The foundation of mathematical calculation model needs soil layer data in exploration report, therefore prospecting parameter is used as the number of all data
Being according to one of basis must be complete complete；The foundation of mathematical calculation model, which is removed, needs to know the soil parameters in exploration report
Outside, in addition it is also necessary to know the design data of foundation ditch, such as foundation ditch plane design drawing, profile, excavation flow and supporting construction parameter,
More accurate so as to the model of foundation, excavating condition more conforms to live actual condition, and the result of calculating more approaches monitoring
Data；Third party monitoring data must be complete, is primarily due to parametric inversion and relies primarily on the data of actual monitoring carry out inverting and obtain
To the Soil Parameters for more approaching actual state, in addition, according to third party monitoring point floor plan, in numerical computations
Corresponding monitoring point is set, makes the displacement of numerical computations with the displacement of actual monitoring there is corresponding locus to sit
Mark, makes whole flow process have invertibility.
1.2) engineering project must include generic Rock And Soil
The model parameter otherness of various soil mass is larger, therefore need to set up between prospecting and the analog parameter of the similar soil body
Relation, this requires required Engineering Projects to contain generic Rock And Soil simultaneously.
1.3) engineering project site must be areal region
The congener soil body physico-mechanical properties of different regions is there is also certain otherness, therefore it is required that required work
Journey case must be areal.
2) processing of pit retaining monitoring data, including：
2.1) error analysis
In the work progress of the Geotechnical Engineerings such as excavation of foundation pit, the Monitoring Data such as displacement generally comprises many factors
Influence, obtained displacement measuring value has error, needs to carry out at data it before back analysis is carried out using displacement data
Reason.
Key element Producing reason and property are measured according to such as displacement etc., can say that error is divided into systematic error, random error
With the class of human error three：Systematic error, random error and human error.
A) systematic error
The reason for systematic error is due to some fixed is caused, and is characterized in regular all the time in whole measurement process
Exist, its absolute value and symbol keep it is constant or it is regular vary, be mainly derived from the following aspects：Method
Error, instrumental error, environmental error, operating error and subjective error.
Systematic error can be summarized changing rule, take corresponding measure to reduce in the measurements by searching related causes
Error, or measurement result is modified in data processing.
B) random error
Random error is caused by some random accidentalia.If carrying out substantial amounts of measurement, certain statistics can be met
Rule, it is considered that obedience is just distributed very much.The reason for producing random error has the side such as measuring instrument, measuring method and environmental condition
The fluctuation in face, such as supply voltage, environment temperature, the minor fluctuations of humidity and air pressure, magnetic interference, the minor variations of instrument, behaviour
Make minute differences on human users etc..Random error is unavoidable in the measurements.It is difficult often area in actual tests
Divide random error and systematic error, therefore many errors are all the combinations of this two classes error.
C) human error
Human error is due to that testing crew is careless, the error that reason causes such as does not handle affairs, such as reads by operational procedure
Wrong meter dial, record and calculating mistake.Human error prevailing value is larger, and is not often substantially inconsistent with practical work.
In addition, when using Monitoring Data with base pit engineering numerical simulation be combined progress parametric inversion when, if choosing
The constitutive relation taken can not embody the substantive characteristics of rock soil medium deformation completely, also lead to result of calculation actual with base pit engineering
Deformation values have differences, and we are for the time being referred to as " essential error "；Other numerical model also can do many to Practical Project
Simplify, equally numerical simulation result can be made to be had differences with practical distortion value.
2.2) Analysis on monitoring data
Because monitoring weekly does not record detailed excavation of foundation pit progress, i.e., without clearly embodiment excavation of foundation pit depth and prison
Survey the corresponding relation of data, it is therefore desirable to careful analysis is carried out to original Monitoring Data, comprehensive considering various effects are arranged
Go out can be used for parameters for numerical simulation inverting, the different cutting depths of correspondence Monitoring Data information.
Monitoring Data processing：Because foundation ditch area is big, construction time limit for length, across autumn and winter spring and summer, centre experienced shut-down, day
The influence (experience frost heave and for several times heavy rain) of the factors such as gas change, all can produce significant impact to foundation pit deformation.This
Outside, due to detecting instrument in itself and the influence that is brought to Monitoring Data of artificial origin is also very important.At the same time, to foundation ditch
Engineering carries out numerical simulation and only accounts for influenceing the essential reason of foundation pit deformation, including the engineering geology bar residing for base pit engineering
Part, hydrogeologic condition, foundation ditch form, support structure design, digging process etc.；According to this structure selected in numerical procedure
Relational model, generally can also ignore the influence that creep, temperature change etc. are produced to foundation pit deformation.Therefore, pit retaining monitoring data are needed
Pre-processed, it is impossible to be directly used in numerical simulation back analysis.How Monitoring Data is reasonably handled, external action of forgoing
Factor, obtains the data message of reflection foundation pit construction process nature deformation rule, is the problem of needing serious analysis.
2.3) Monitoring Data is handled
Measured data by site operation environment, working procedure, accuracy of instrument, Changes in weather, artificial disturbance etc. it is numerous because
The influence of element, and it is to be carried out under preferable environment to simulate, the result that both settle accounts can produce error unavoidably, to control as far as possible
Error rejects some singular points, it is necessary to contrasted and screened to Monitoring Data, what condition was met, and measured data can be carried out
Fitting, obtains relatively reasonable measured data.
3) mathematical calculation model of standard is set up, following standardisation requirements are met：
3.1) foundation of numerical model
The model of foundation is the part of engineering entity, between numerical model is set up, and the data of monitoring need to be carried out
Induction and conclusion, selection monitoring type is more, position that is relatively complete, relatively completing, and the surrounding enviroment at simultaneous selection modeling position are more simple
Single, physical form is simple, rule, if being easy to modeling, condition satisfaction, preferably sets up complete mathematical calculation model.
The size of model takes 3~5 times of foundation depth, and model bottom is fixed boundary, and its X, Y, Z-direction position are constrained respectively
Move；It is the scope of freedom at the top of model, is not provided with constraint；Model constrains the displacement of the horizontal direction respectively in two horizontal directions.
The model meshes of foundation are divided and also there are certain requirements, and density need to divide appropriate, and structure is employed during modeling
Grid at unit, construction unit need to divide comparatively dense some because at most there is a construction unit section among a cell cube
Point.While excavating the key positions such as step and foundation ditch bottom when modeling, interface should be set, prevent excavation is not in place from being produced to result of calculation
Raw certain influence.
Cast-in-situ bored pile is included in the supporting construction of foundation ditch case, and the stress form of cast-in-situ bored pile and underground are continuous
Wall is close, and by the equal principle of bending rigidity, by cast-in-situ bored pile building enclosure, this is counted as certain thickness diaphram wall,
Equivalent formula such as formula 1.
In formula：H is equally connects wall thickness, and D is cast-in-situ bored pile diameter, and t is the clear distance of stake.
Ground-connecting-wall uses solid element, and contact surface is set between ground-connecting-wall and the soil body.
3.2) classification of constitutive model
Conventional constitutive model is divided into the bullet of 3 major classes, elastic class model, bullet-ideal plasticity class model and strain hardening class
Plasticity model.Elastic model is due to that can not reflect the plastic property of the soil body, preferably simulation active earth pressure and can not be broken ground
Pressure is thus not suitable for the analysis of excavation of foundation pit.The MC models and DP models of bullet-ideal plasticity are due to past using single rigidity
It is past to cause very big hole bottom resilience, it is difficult to while providing soil deformation after rational deformation of wall and wall.It can consider that bury should
Become hardening characteristics, the difference that loading and off-load can be distinguished and its rigidity dependent on stress history and the hardening class model of stress path
Such as MCC models and HS models, soil deformation situation after relatively reasonable deformation of wall and wall can be provided simultaneously.Under sensitive environment
Design of Excavation Project needs the deformation of the soil body after critical concern wall, from meeting requirement of engineering and convenient and practical angle,
It is proposed with the excavation of foundation pit numerical analysis under the sensitive environment of MCC and HS models progress.According to the characteristics of model itself, Ke Yi great
Cause to judge applicability of the various constitutive models in foundation ditch numerical value excavates analysis, refer to Fig. 1.
3.3) selection of constitutive model and its determination of initial parameter values
According to step 2.2), mainly using modified Cam-clay (MCC models), sandy gravel uses mole coulomb model (MC
Model).Modified Cam-clay is needed in 4 model parameters, i.e. v-lnp' planes in slope λ, the v-lnp' plane of normal consolidation line
The slope M of CSL lines, Poisson's ratio v (or shear modulus G) on slope κ, p'-q' faces of resilience line.Wherein λ and κ can be real according to consolidation
Test and tried to achieve respectively by formula 2 and formula 3.
C in formulacAnd CsRespectively native cake compressibility and swelling index.M can be obtained according to triaxial compression test by formula 4.
In formulaThe effective angle of inner friction obtained for triaxial test.
In addition modified Cam-clay is still needed 2 state parameters, i.e. initial void ratio e0With preconsolidation pressure p0。
The major parameter of mole coulomb model is：Natural density γ, cohesive strength c, internal friction angleModulus of pressure E, Poisson
Compare v.
3.4) generation of sample parameter
Obtain after initial parameter, carry out initial numerical computations, and contrast with measured result.It is determined that needing the soil of inverting
Layer and inverting Soil Parameters, then on the basis of calculating parameter, with certain multiple extended parameter (parameter of expansion
Result of calculation should be able to include the result of actual monitoring), then the extended parameter of inverting soil layer is subjected to permutation and combination, form sample ginseng
Number, finally substitutes into computation model by the parameter of sample, is respectively calculated, enters the result of calculation of sample parameter after the completion of calculating
Row is arranged.In addition also need to randomly select several groups of test parameters being located between sample parameter (for testing neutral net mould
Type), carry out numerical computations and arrange its result.
4) Parameter Inversion Model is set up, Inversion Calculation is carried out, including：
4.1) qualifications that parametric inversion is calculated
The calculating means of parametric inversion use the BP neural network of genetic algorithm optimization, and (its algorithm flow refers to figure
2), there is certain requirement to sample using this inverse model：A) number of species of measured data must be not less than soil body inverted parameters
Number of species；B) result that sample parameter is calculated must can cover the data of actual measurement.Therefore during sample is generated, inverting
Parameter to choose rationally, meet the requirement of parametric inversion system, otherwise result of calculation can produce larger error.
4.2) calculation process of parametric inversion mainly includes：
A) input, the design of output layer
By step 2.4) numerical result substitute into neural network model, using Soil Parameters as input layer, by the soil body
The numerical result that parameter is obtained is as output layer.
B) determination of the number of plies is implied
The determination of the number of hidden nodes does not have clear and definite analytic expression to represent, need to be determined according to practical experience.Because not being bigger
It is better, nor the smaller the better, it is therefore desirable to choose a moderate number.For single hidden layer network, in practical application, the upper bound
For number of training, lower bound is output unit number.If using pyramid rule, if that is, only one layer hidden layer network, hidden
Node layer number can useWherein n and m are respectively input and output layer nodes.Also have using 2n+1 initial hidden nodes
Number, specific number is determined by tentative calculation.
C) normalization of training sample
Normalization is using, from tape function mapminmax, the form of this function normalization is with maximum in one group in MATLAB
Value subtracts minimum value as denominator, subtracts minimum value as molecule using normalized number is wanted, you can obtain the normalized value of the number.
The normalized of data is that all data are converted to number between [0,1], and the purpose is to cancel quantity between each dimension data
Level difference, it is to avoid because inputoutput data order of magnitude difference is larger and causes neural network forecast error larger.
D) neutral net is tentatively set up
BP neural network is tentatively set up using the newff functions tool box in MATLAB, concrete form is as follows：
Net=newff (a, b, hiddennum)；
A represents the input sample after the normalization of numerical simulation calculation result；
B represents the output sample after the normalization of numerical simulation Soil Parameters；
Hiddennum represents node in hidden layer.
E) genetic algorithm parameter is initialized
The parameter contained in genetic algorithm is initialized, parameter includes：Iterations, population scale (chromosome
Number), crossover probability, mutation probability.In Genetic Algorithm Optimized Neural Network problem, chromosome is exactly the weights by neutral net
Constituted with threshold value.
F) initialization of population
By each chromosome initialization in population, will every group of weights and threshold value assign initial value, method is used in MATLAB
Rands functions initialized.
G) calculating of chromosome fitness function
The initial weight and threshold value of BP neural network are obtained according to individual, is trained with training data and system is predicted after BP neural network
System output, using the Error Absolute Value and E between prediction output and desired output as ideal adaptation angle value F, calculation formula is shown in formula 5.
In formula：N is network output node number；yiFor the desired output of i-th of node of BP neural network；OiFor i-th of node
Prediction output；K is coefficient.
The chromosome (weights and threshold value) of every group of initialization is assigned the network (net) tentatively set up, after normalization
Numerical simulation calculation result carries out the training of network as the Soil Parameters after input sample and normalization as output sample, and
Tested, using the systematic error of test as homologue (weights and threshold value) fitness function value, because error is got over
It is small better, so the minimum chromosome (weights and threshold value) of systematic error is optimum individual in this generation population.Record fitness letter
Number optimal individual and its fitness function.
H) chromosome is selected
Chromosome is selected using wheel disc pronunciation, i.e. the selection strategy based on fitness ratio, each individual i choosing
Select probability PiSee formula 6.
In formula：FiFor individual i fitness value, because fitness value is the smaller the better, so to adapting to before individual choice
Angle value asks reciprocal；K is coefficient；N is population at individual number.
I) chiasma
Because individual uses real coding, so crossover operation method uses real number interior extrapolation method, k-th of chromosome akWith
L chromosome alSee formula 7 in the crossover operation methods of j.
In formula：B is the random number between [0,1].
J) chromosomal variation
Choose j-th of gene a of i-th of individualijEnter row variation, mutation operation method is shown in formula 8.
In formula：amaxFor gene aijThe upper bound；aminFor gene aijLower bound；F (g)=r2(1-g/Gmax)2；r2For one
Random number；G is current iteration number of times；GmaxFor maximum evolution number of times；R is the random number between [0,1].
K) chromosome fitness value is updated
The fitness function value of chromosome of new generation is calculated again, and method is as (g) step.Each neozoic dye
Optimal individual is compared in the fitness value of colour solid and the previous generation, updates optimum individual and its fitness value.Then circulation changes
Generation, until reaching the iterations of setting.
L) assign neutral net by optimal chromosome and carry out neural metwork training
Neutral net finally is assigned using optimal chromosome (weights, threshold value) as the initial weight threshold value of neutral net, and
The result of calculation of the numerical simulation after normalization is regard as output as the parameter of the simulation soil body after input sample and normalization
Sample carries out the training of network, and then the mapping set up between the data result of numerical simulation calculation and numerical simulation Soil Parameters
Relation.
M) neural network model is tested
By step 2.4) in test parameter substitute into the neural network model that step l) sets up and carry out network test, calculate
The error amount of each output variable in each test sample.If training result is unsatisfactory for requiring, genetic algorithm ginseng is adjusted again
Number, re-starts the optimization of network weight threshold value until meeting error requirements.
N) calculating parameter inverting
Actual monitoring data are normalized, the measured data after normalization is substituted into the nerve that step l) is set up
Network model, inverting obtains the corresponding simulation Soil Parameters of measured displacements.
5) relation between prospecting and analog parameter is set up：
Multiple base pit engineerings are respectively calculated according to above-mentioned steps and obtain simulation soil body ginseng corresponding by measured displacements
Number, then again using the prospecting parameter of different base pit engineerings as input layer, the simulation Soil Parameters that inverting is obtained as output layer,
Neutral net is substituted into be trained, and then the mapping relations reconnoitred between analog parameter, and look for an Engineering Projects to carry out
Checking.
Embodiment 1：
The specific embodiment for setting up open-cut foundation ditch engineering investigation and analog parameter relational approach based on the present invention include with
Lower step：
Step 1：According to monitoring plan, Monitoring Data (pile head settlement, ground settlement, the stake top of each monitoring point are extracted
Horizontal displacement etc.) (Sa1…Sf1), and prospecting parameter (a1…f1), Monitoring Data is analyzed, some singular points are rejected, so
Monitoring Data is handled using methods such as fittings afterwards, the Monitoring Data (S of key point is extracteda2…Sf2)。
According to the difference of support form, the appearance form of pile lateral displacement is also different, and such as pile lateral displacement is presented
Measured data, can be fitted to exponential function, then take the Monitoring Data at its stake top position by " rim of a cup " type；If measured data is in
Existing " bulging " type, can be fitted to just too distribution function, take the Monitoring Data at its maximum displacement position；Which kind of letter be specifically fitted to
Number, also needs dynamically to be adjusted and value according to the engineering characteristic of base pit engineering and the changing rule of Monitoring Data.
Step 2：According to the monitoring plan and Monitoring Data of step 1, selection surrounding enviroment are relatively easy, monitoring species phase
To more, the Monitoring Data relatively simple position of complete, foundation ditch shape relatively, according to the actual formation condition of foundation ditch, it is based on
FLAC3D5.0 numerical softwares set up the excavation of foundation pit model of standard with 3~5 times of foundation depth, and set its boundary condition.
Step 3：Simulation initial parameter (N is drawn by prospecting relevant parameter or experiencea2…Nf2) (including cohesive strength, interior friction
Angle, modulus of compressibility, Poisson's ratio, natural density, the slope λ of normal consolidation line, it is elastic expansion line slope κ, constant of friction M, initial
Void ratio e0With preconsolidation pressure p0) and calculated, according to result of calculation, around simulation initial parameter (Na2…Nf2) be multiplied by
Certain coefficient draws multigroup sample parameter (Nai…Nfi), sample parameter is then substituted into numerical model respectively and calculated, is obtained
Go out its result of calculation (Sai…Sfi).Several groups of test sample parameter (n are randomly selected between sample parameterai…nfi), and enter respectively
Row numerical computations, draw its result of calculation (sai…sfi)。
FLAC3D numerical computations, (are computed repeatedly, model is constant, only modification soil body ginseng due to needing to calculate multigroup sample
Number), therefore the command file " .txt " of batch processing need to be write, and preserve result of calculation " .sav " form text of different sample parameters
Part (its name must modify, otherwise will capped), for convenience of subsequent treatment, it would be desirable to result of calculation be output as " .xls "
Excel table format files, above file is placed in same file and is named.
Parametric inversion of the present invention calculates the BP neural network for using genetic algorithm optimization, and program code is to be based on
Matlab carries " .m " formatted file write in artificial neural network tool box, and the data message such as input and output variable is protected
Save as " .mat " formatted file.So if to perform order being fitted without on the computer of matlab softwares, it is necessary first to
By matlab software installations on computers, need to open " .m " file in matlab during operation neutral net, adjust
With the data in " .mat " file, operation calculates and preserves network, and network trust system is in " .m " file.
Step 4：The result of calculation that step 3 is obtained substitutes into neural network model, by soil body sample parameter (Nai…Nfi) make
For input layer, its corresponding simulation result of calculation (Sai…Sfi) as output layer, node in hidden layer can useWherein n and
M is respectively input and output layer nodes, can also use 2n+1 initial the number of hidden nodes, and specific number is true by tentative calculation
It is fixed.Population in Genetic Algorithms scale, iterations, intersection and mutation probability are set, neural network model is then run, sets up the soil body
Mapping relations between sample parameter and simulation result of calculation.
Step 5：By test sample parameter (nai…nfi) substitute into soil body sample parameter and simulation calculating knot that step 4 is set up
Mapping relations between fruit, the neural network model that operating procedure 4 is preserved is calculated, and draws the corresponding nerve net of test sample
Network result of calculation (Lai…Lfi), comparative analysis test sample parameter (nai…nfi) corresponding numerical simulation calculation result (sai…
sfi) and neural computing result (Lai…Lfi) error, meet require, then carry out next step.
Step 6：By actual monitoring data (Sa1…Sf1) substitute into step 4 set up neural network model, run and counted
Calculate, and then draw by the Soil Parameters (M of the corresponding simulation of measured dataa1…Mf1), each open-cut foundation ditch engineering is according to above-mentioned
Flow is respectively calculated.The prospecting parameter of all base pit engineering cases is (ai…fi), it is anti-according to above-mentioned steps by measured data
Drill the Soil Parameters (M for calculating obtained simulationai…Mfi)
Step 7：Will prospecting parameter (ai…fi) as the input layer of neutral net, obtained by measured data Inversion Calculation
Soil Parameters (the M of simulationai…Mfi) as the output layer of neutral net, according to step 4, artificial neural network and genetic algorithms are set
Correlation computations parameter, operation neural network model is calculated, and then is drawn between open-cut foundation ditch engineering investigation and analog parameter
Mapping relation.
Step 8：With a new open-cut foundation ditch Engineering Projects, and the prospecting parameter of the base pit engineering case is extracted, be updated to
The open-cut foundation ditch engineering investigation that step 7 is set up show that its is right with being calculated in the neural network model of analog parameter relation
The Soil Parameters for the simulation answered, then set up the numerical model of new base pit engineering, and the Soil Parameters of the simulation drawn are substituted into
Calculated in numerical model, its result of calculation and measured result are contrasted, analyze whether its error meets requirement.If not
It is full to require, then using the method for amendment numerical model, the constitutive model using higher level, amendment neural network parameter etc., again
Flow according to above-mentioned steps 2 to step 7 calculate and required until meeting.
If subsequently having new base pit engineering case again, can again it be counted according to the flow of above-mentioned steps 1 to step 8
Calculate, constantly the relation between the set up open-cut foundation ditch engineering investigation of amendment and analog parameter.
Claims (4)
1. a kind of method for setting up open-cut foundation ditch engineering investigation and analog parameter relation, it is characterised in that comprise the following steps：
Step 1, the Monitoring Data for gathering base pit engineering, are fitted to it according to its changing rule, are used as the base of Inversion Calculation
Plinth data；
Step 2, the mathematical calculation model for setting up standard, choose suitable soil constitutive model；Extract the original prospecting ginseng of soil layer
Number, the initial soil calculation parameter of numerical computations is converted to this；Based on initial soil calculation parameter, numerical value meter is expanded to
The sample parameter and test sample parameter of calculation；Sample parameter and test sample parameter are subjected to numerical computations respectively；
Step 3, the BP neural network that sample parameter and numerical result are substituted into genetic algorithm optimization are trained, and set up soil
Mapping relations between body parameter and numerical result, and test neural network model with test sample parameter；Will be actual
Monitoring Data substitutes into Soil Parameters and the relation of result of calculation tries to achieve the corresponding Soil Parameters of measured displacements；To multiple base pit engineerings
Case is respectively calculated, by the simulation corresponding with the measured displacements that inverting is obtained of the prospecting parameter of multiple base pit engineering cases
Soil Parameters substitute into the BP neural network of genetic algorithm optimization, set up the relation of open-cut foundation ditch engineering investigation and analog parameter.
2. the method as claimed in claim 1 for setting up open-cut foundation ditch engineering investigation and analog parameter relation, it is characterised in that step
Monitoring Data in rapid 1 includes：Pile head settlement, ground settlement, Horizontal Displacement, pile lateral displacement.
3. the method as claimed in claim 1 for setting up open-cut foundation ditch engineering investigation and analog parameter relation, it is characterised in that step
Original prospecting parameter in rapid 2 includes：Cohesive strength, internal friction angle, modulus of compressibility, Poisson's ratio, natural density, natural moisture content,
Proportion, void ratio, liquid limit, plasticity index, cake compressibility, swelling index.
4. the method as claimed in claim 3 for setting up open-cut foundation ditch engineering investigation and analog parameter relation, it is characterised in that step
Rapid 3 specifically include following steps：
Step 3.1：The result of calculation that step 2 is obtained substitutes into neural network model, by soil body sample parameter (Nai…Nfi) conduct
Input layer, its corresponding simulation result of calculation (Sai…Sfi) as output layer, node in hidden layer can useWherein, n and m
Respectively input and output layer nodes, using 2n+1 initial the number of hidden nodes, set Population in Genetic Algorithms scale, iteration time
Number, intersection and mutation probability, then run neural network model, set up reflecting between soil body sample parameter and simulation result of calculation
Penetrate relation；
Step 3.2：By test sample parameter (nai…nfi) substitute into soil body sample parameter and simulation result of calculation that step 3.1 is set up
Between mapping relations, operating procedure 3.1 preserve neural network model calculated, draw the corresponding nerve net of test sample
Network result of calculation (Lai…Lfi), comparative analysis test sample parameter (nai…nfi) corresponding numerical simulation calculation result (sai…
sfi) and neural computing result (Lai…Lfi) error, meet require, then carry out next step；
Step 3.3：By actual monitoring data (Sa1…Sf1) substitute into step 3.1 set up neural network model, run and calculated,
And then draw by the Soil Parameters (M of the corresponding simulation of measured dataa1…Mf1), each open-cut foundation ditch engineering is according to above-mentioned stream
Journey is respectively calculated.The prospecting parameter of all base pit engineering cases is (ai…fi), by measured data according to above-mentioned steps inverting
Calculate the Soil Parameters (M of obtained simulationai…Mfi)；
Step 3.4：Will prospecting parameter (ai…fi) as the input layer of neutral net, the mould obtained by measured data Inversion Calculation
Soil Parameters (the M of planai…Mfi) as the output layer of neutral net, according to step 3.1, artificial neural network and genetic algorithms are set
Correlation computations parameter, operation neural network model is calculated, and then is drawn between open-cut foundation ditch engineering investigation and analog parameter
Mapping relation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710438743.0A CN107315862B (en) | 2017-06-12 | 2017-06-12 | Method for establishing open cut foundation pit engineering investigation and simulation parameter relationship |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710438743.0A CN107315862B (en) | 2017-06-12 | 2017-06-12 | Method for establishing open cut foundation pit engineering investigation and simulation parameter relationship |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107315862A true CN107315862A (en) | 2017-11-03 |
| CN107315862B CN107315862B (en) | 2020-11-27 |
Family
ID=60183745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710438743.0A Active CN107315862B (en) | 2017-06-12 | 2017-06-12 | Method for establishing open cut foundation pit engineering investigation and simulation parameter relationship |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107315862B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109870376A (en) * | 2019-02-03 | 2019-06-11 | 浙江大学 | A method of based on nano impress and numerical simulation inverting rock forming mineral parameter |
| CN111080020A (en) * | 2019-12-23 | 2020-04-28 | 中山大学 | Robustness evaluation method and device for drilling arrangement scheme |
| CN111159935A (en) * | 2019-12-11 | 2020-05-15 | 同济大学 | BP neural network parameter calibration method based on LHS |
| CN111705808A (en) * | 2020-06-30 | 2020-09-25 | 河海大学 | Soil body parameter dynamic inversion analysis method suitable for suspended deep foundation pit engineering |
| CN112347670A (en) * | 2020-10-26 | 2021-02-09 | 青海大学 | Rockfill material creep parameter prediction method based on neural network response surface |
| CN112459080A (en) * | 2020-11-16 | 2021-03-09 | 上海市机械施工集团有限公司 | Intelligent adjusting method for steel support axial force based on bp neural network |
| CN113094948A (en) * | 2021-03-29 | 2021-07-09 | 天津大学 | Intelligent early warning cloud system for geotechnical underground engineering |
| CN113158586A (en) * | 2021-05-26 | 2021-07-23 | 华能新能源股份有限公司 | Wind power plant numerical simulation method and device combined with neural network algorithm and storage medium |
| CN113378276A (en) * | 2021-06-18 | 2021-09-10 | 北方工业大学 | Composite foundation intelligent design method based on genetic algorithm and gene expression programming |
| CN113946899A (en) * | 2021-11-10 | 2022-01-18 | 广东粤海珠三角供水有限公司 | Slurry shield tunneling parameter prediction method based on real-time geological information |
| CN116579255A (en) * | 2023-07-12 | 2023-08-11 | 西南交通大学 | CPTU model test-based intelligent identification method for correcting Cambridge model parameters |
| CN118133672A (en) * | 2024-03-18 | 2024-06-04 | 北京市生态地质研究所 | Foundation pit supporting performance evaluation method and system based on data analysis |
| CN118133635A (en) * | 2024-05-07 | 2024-06-04 | 深地科学与工程云龙湖实验室 | Tunnel contact channel deformation control method, system, equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8285659B1 (en) * | 2009-08-18 | 2012-10-09 | The United States of America as represented by the Administrator of the National Aeronautics & Space Administration (NASA) | Aircraft system modeling error and control error |
| CN103198215A (en) * | 2013-03-20 | 2013-07-10 | 辽宁工程技术大学 | Method for predicting subsidence of soil outside pit based on differential evolution support vector machine |
| CN103205972A (en) * | 2013-03-19 | 2013-07-17 | 辽宁工程技术大学 | Method for analyzing relationship between deformation of foundation pit and ground subsidence outside of foundation pit |
| CN104077451A (en) * | 2014-07-03 | 2014-10-01 | 中铁上海工程局集团有限公司 | Deep soft soil metro foundation pit soil body parameter inversion analyzing method |
| CN104965969A (en) * | 2015-04-20 | 2015-10-07 | 广西大学 | Inversion method for surrounding rock mechanical parameters of large cavern group |
| CN106400813A (en) * | 2016-10-28 | 2017-02-15 | 北京城建勘测设计研究院有限责任公司 | Foundation pit support system and construction method thereof |
-
2017
- 2017-06-12 CN CN201710438743.0A patent/CN107315862B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8285659B1 (en) * | 2009-08-18 | 2012-10-09 | The United States of America as represented by the Administrator of the National Aeronautics & Space Administration (NASA) | Aircraft system modeling error and control error |
| CN103205972A (en) * | 2013-03-19 | 2013-07-17 | 辽宁工程技术大学 | Method for analyzing relationship between deformation of foundation pit and ground subsidence outside of foundation pit |
| CN103198215A (en) * | 2013-03-20 | 2013-07-10 | 辽宁工程技术大学 | Method for predicting subsidence of soil outside pit based on differential evolution support vector machine |
| CN104077451A (en) * | 2014-07-03 | 2014-10-01 | 中铁上海工程局集团有限公司 | Deep soft soil metro foundation pit soil body parameter inversion analyzing method |
| CN104965969A (en) * | 2015-04-20 | 2015-10-07 | 广西大学 | Inversion method for surrounding rock mechanical parameters of large cavern group |
| CN106400813A (en) * | 2016-10-28 | 2017-02-15 | 北京城建勘测设计研究院有限责任公司 | Foundation pit support system and construction method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 宋春晖: "地铁车站深基坑土体参数动态反演分析及变形预测", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109870376A (en) * | 2019-02-03 | 2019-06-11 | 浙江大学 | A method of based on nano impress and numerical simulation inverting rock forming mineral parameter |
| CN109870376B (en) * | 2019-02-03 | 2020-10-23 | 浙江大学 | Rock mineral parameter inversion method based on nano indentation and numerical simulation |
| CN111159935A (en) * | 2019-12-11 | 2020-05-15 | 同济大学 | BP neural network parameter calibration method based on LHS |
| CN111080020A (en) * | 2019-12-23 | 2020-04-28 | 中山大学 | Robustness evaluation method and device for drilling arrangement scheme |
| CN111080020B (en) * | 2019-12-23 | 2023-03-31 | 中山大学 | Robustness evaluation method and device for drilling arrangement scheme |
| CN111705808A (en) * | 2020-06-30 | 2020-09-25 | 河海大学 | Soil body parameter dynamic inversion analysis method suitable for suspended deep foundation pit engineering |
| CN112347670B (en) * | 2020-10-26 | 2022-09-27 | 青海大学 | Rockfill material creep parameter prediction method based on neural network response surface |
| CN112347670A (en) * | 2020-10-26 | 2021-02-09 | 青海大学 | Rockfill material creep parameter prediction method based on neural network response surface |
| CN112459080B (en) * | 2020-11-16 | 2022-04-15 | 上海市机械施工集团有限公司 | Intelligent adjusting method for steel support axial force based on bp neural network |
| CN112459080A (en) * | 2020-11-16 | 2021-03-09 | 上海市机械施工集团有限公司 | Intelligent adjusting method for steel support axial force based on bp neural network |
| CN113094948A (en) * | 2021-03-29 | 2021-07-09 | 天津大学 | Intelligent early warning cloud system for geotechnical underground engineering |
| CN113158586A (en) * | 2021-05-26 | 2021-07-23 | 华能新能源股份有限公司 | Wind power plant numerical simulation method and device combined with neural network algorithm and storage medium |
| CN113378276A (en) * | 2021-06-18 | 2021-09-10 | 北方工业大学 | Composite foundation intelligent design method based on genetic algorithm and gene expression programming |
| CN113946899A (en) * | 2021-11-10 | 2022-01-18 | 广东粤海珠三角供水有限公司 | Slurry shield tunneling parameter prediction method based on real-time geological information |
| CN116579255A (en) * | 2023-07-12 | 2023-08-11 | 西南交通大学 | CPTU model test-based intelligent identification method for correcting Cambridge model parameters |
| CN116579255B (en) * | 2023-07-12 | 2023-09-05 | 西南交通大学 | CPTU model test-based intelligent identification method for correcting Cambridge model parameters |
| CN118133672A (en) * | 2024-03-18 | 2024-06-04 | 北京市生态地质研究所 | Foundation pit supporting performance evaluation method and system based on data analysis |
| CN118133635A (en) * | 2024-05-07 | 2024-06-04 | 深地科学与工程云龙湖实验室 | Tunnel contact channel deformation control method, system, equipment and storage medium |
| CN118133635B (en) * | 2024-05-07 | 2024-08-16 | 深地科学与工程云龙湖实验室 | Tunnel contact channel deformation control method, system, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107315862B (en) | 2020-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107315862A (en) | A kind of method for setting up open-cut foundation ditch engineering investigation and analog parameter relation | |
| Robertson | Cone penetration test (CPT)-based soil behaviour type (SBT) classification system—an update | |
| CN106952003A (en) | High Ground Stress Areas beded rock mass underground rock cavern Failure type Forecasting Methodology | |
| CN108842821B (en) | Calculation method for reasonable buried depth of submarine tunnel constructed by drilling and blasting method | |
| CN109752262B (en) | Method for determining dynamic shear modulus parameter of covering soil mass based on in-situ relative density | |
| Xing et al. | Comprehensive monitoring of talus slope deformation and displacement back analysis of mechanical parameters based on back-propagation neural network | |
| Al-saffar et al. | Prediction of soil's compaction parameter using artificial neural network | |
| CN115099111B (en) | Shield construction ground surface settlement prediction method based on machine learning and finite element | |
| CN113158315A (en) | Rock-soil body parameter three-dimensional non-stationary condition random field modeling method based on static cone penetration data | |
| Johari et al. | Stochastic nonlinear ground response analysis: A case study site in Shiraz, Iran | |
| Liu et al. | Inverse procedure for calibrating parameters that control land subsidence caused by subsurface fluid withdrawal: 2. Field application | |
| CN109540738B (en) | Method for determining in-situ relative density of deep overburden soil body by considering soil layer types | |
| Gillins | Mapping the probability and uncertainty of liquefaction-induced ground failure | |
| CN116776707A (en) | Foundation pit structure digital numerical analysis method based on geological drilling | |
| Tang et al. | Variability of predictions in geotechnics | |
| Yang et al. | Computation and analysis of high rocky slope safety in a water conservancy project | |
| Pierce et al. | Synthetic rock mass applications in mass mining | |
| CN112064617A (en) | Soil-stone mixture foundation quality detection method | |
| Mangiola et al. | An application of Stochastic Site Response Analysis | |
| Sarker et al. | MATLAB modeling of SPT and grain size data in producing soil-profile | |
| Sharma et al. | Seismic Ground Response Analysis using Continuum Approach | |
| US20240094419A1 (en) | Seismic quantitative prediction method for shale toc based on sensitive parameter volumes | |
| Hajian et al. | Prior applications of neural networks in geophysics | |
| Mikkelsen et al. | Numerical assessment on pile stability in liquefiable soil | |
| CN117648747A (en) | Oversized foundation pit stability analysis method considering soil parameter space variability |
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 |