CN114547951B - Dam state prediction method and system based on data assimilation - Google Patents
Dam state prediction method and system based on data assimilation Download PDFInfo
- Publication number
- CN114547951B CN114547951B CN202210432633.4A CN202210432633A CN114547951B CN 114547951 B CN114547951 B CN 114547951B CN 202210432633 A CN202210432633 A CN 202210432633A CN 114547951 B CN114547951 B CN 114547951B
- Authority
- CN
- China
- Prior art keywords
- dam
- model
- data
- assimilation
- displacement
- 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.)
- Active
Links
Images
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
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/13—Architectural design, e.g. computer-aided architectural design [CAAD] related to design of buildings, bridges, landscapes, production plants or roads
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping
Abstract
The invention discloses a dam state prediction method and system based on data assimilation, and belongs to the technical field of dam state prediction. The existing dam state prediction method has the advantages that the calculation accuracy extremely depends on the correct description of the concrete material constitutive model, model parameters can be generally obtained only from prototype observation or twin experiments, and the parameters are determined and calibrated more complexly. According to the method, the nonlinear evolution characteristic of the concrete material along with time is fully considered, a data assimilation model is constructed by using the displacement of a dam body virtual displacement field and the displacement of an actual measuring point of the dam at the same time, and the linear elasticity constitutive parameter of the concrete material of the dam is subjected to displacement inverse analysis; and observation data is introduced into the data assimilation model for model updating, the influence of data errors on the model is fully considered, and the evolution direction of the data assimilation model is adjusted in real time, so that the estimation precision of the data assimilation model is improved, the prediction capability of the prediction model is effectively improved, and important reference data can be provided for real-time early warning analysis of a dam.
Description
Technical Field
The invention relates to a dam state prediction method and system based on data assimilation, and belongs to the technical field of dam state prediction.
Background
The dam is a special building comprising a reservoir and a hydropower station, plays roles in storing water, preventing flood, generating electricity and the like, and has remarkable social and economic benefits. However, the dam projects in China have the characteristics of a plurality of points, wide areas and large quantity, once serious accidents such as dam break occur, immeasurable damage is caused to the downstream masses and the environment, so that the construction quality and the maintenance and operation safety of the dam projects must be ensured.
However, dam engineering has the characteristics of large building scale, complex design structure, long construction service period, various operating environment changes and the like, and the main body material concrete also has the physical properties of dynamic changes such as hydration heat release, drying shrinkage, creep and the like, and all the factors can influence the displacement, stress and strain states of the dam. Although dam state monitoring methods and systems are relatively complete, most of them are safety threshold determination methods based on monitoring, and in summary, the existing dam state prediction methods have the following disadvantages:
a prediction method based on a statistical model, such as a neural network, a regression model, a genetic algorithm and the like, belongs to an empirical model or an empirical analysis method, ignores dynamic structural mechanics evolution in the process of dam bearing and temperature rise and temperature drop, lacks of a real physical model support and explanation basis, and cannot accurately predict situations which do not occur in history, such as extreme weather and the like.
According to the prediction method based on simulation analysis, the physical interpretation basis is sufficient, but the dam structure presents a space-time nonlinear behavior, the matrix decomposition and solving calculation scale is large, and the time consumption is long; the calculation accuracy extremely depends on correct description of a concrete material constitutive model, model parameters can be generally obtained only from prototype observation or twin experiments, and the parameters are complex to determine and calibrate. The prediction is difficult to predict in real time and accurately.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a method for carrying out displacement inverse analysis on the linear elastic constitutive parameters of the concrete material of the dam by constructing a data assimilation model by fully considering the nonlinear evolution characteristic of the concrete material along with time; observation data are introduced into the data assimilation model to update the model, the influence of data errors on the model is fully considered, the evolution direction of the data assimilation model is adjusted in real time, the estimation precision of the data assimilation model is improved, the prediction capability of the prediction model is effectively improved, and the dam state prediction method and the dam state prediction system based on data assimilation are simple and practical in scheme.
In order to achieve the above object, a first technical solution of the present invention is:
a dam state prediction method based on data assimilation,
the method comprises the following steps:
the first step is as follows: obtaining the stress and deformation displacement of the dam as the initial physical field of the dam;
the second step is that: performing thermal coupling calculation according to the initial physical field in the first step to obtain a dam body virtual displacement field at a certain moment;
the third step: constructing a data assimilation model by using the dam body virtual displacement field in the second step and the displacement of the actual measuring point of the dam at the same time, and performing displacement inverse analysis on the concrete material linear elasticity constitutive parameters of the dam;
observation data are introduced into the data assimilation model to update the model, and the evolution direction of the data assimilation model is adjusted in real time according to the influence of data errors on the model so as to improve the estimation precision of the data assimilation model;
the fourth step: and calculating material parameters through the data assimilation model in the third step, and constructing a prediction model according to the collected dam monitoring data to predict the dam state.
Through continuous exploration and test, the nonlinear evolution characteristic of the concrete material along with time is fully considered, namely the current situation that the virtual displacement field and the actual displacement field of the concrete material have larger difference along with the change of time is fully considered, and the concrete material linear elasticity constitutive parameters of the dam are subjected to displacement inverse analysis by constructing a data assimilation model; and the observation data is introduced into the data assimilation model to update the model, the influence of data errors on the model is fully considered, and the evolution direction of the data assimilation model is adjusted in real time, so that the estimation precision of the data assimilation model is improved, the prediction capability of the prediction model is effectively improved, and the scheme is simple, practical and feasible.
Furthermore, compared with nonlinear calculation considering creep and hydration heat, the method effectively improves the calculation efficiency, can ensure better accuracy of results obtained through data assimilation calculation in short-term prediction, and can provide important reference data for real-time early warning analysis of the dam.
Furthermore, the certain time can be the current time, the method can obtain a prediction model of the dam formed by the physical field assimilated at the current time, and can predict the short-term physical state of the dam in real time; therefore, the safety monitoring and real-time early warning of the dam are supported, and a basis is provided for daily operation and maintenance, reservoir water level adjustment in extreme weather such as flood, sudden temperature rise and fall, strong snowfall and the like.
As a preferable technical measure:
in the first step, the method for obtaining the stress and the deformation displacement comprises the following steps:
constructing a regular three-dimensional structure hexahedral mesh model based on the design parameters of the dam and the parameters of the elastic material, and establishing a pouring model of each dam segment according to the three-dimensional structure hexahedral mesh model, wherein the pouring model is used for reflecting the hydration heat reaction and the drying shrinkage process of concrete; and (4) according to the pouring model, carrying out simulation calculation on the dam construction process, eliminating horizontal stress between transverse seams of different dam sections, and obtaining the stress and deformation displacement of the dam.
As a preferable technical measure:
the elastic material parameters comprise elastic modulus, Poisson's ratio, thermal expansion coefficient, density, thermal conductivity and specific heat capacity, and are obtained by detecting a concrete sampling test piece of the dam.
As a preferable technical measure:
in the second step, the method for acquiring the virtual displacement field is as follows:
dam monitoring data at a certain time is taken as an open boundary condition, and dam face temperature T is set according to the environment temperature, the water temperature before the dam, the reservoir water level of the dam and the effect of sediment accumulationi Practice ofAs the heat source boundary of the dam, a hexahedral primary grid is taken as a discrete unit;
introducing the dam body temperature H according to the hydration heat reaction and the self-generated heat source existing in the dami In factAnd a correction value thetaiAnd calculating to obtain the virtual temperature field T of the dam at the ith momenti Virtualization；
The virtual temperature field T at the ith momenti VirtualizationThe hexahedral secondary grid mapped to the dam is used as a thermal boundary, and the water level before the dam is set
Zi Practice ofAs the pressure boundary of the dam, only the action of the heat boundary and the pressure boundary is utilized to carry out thermal coupling analysis, and the virtual displacement field T of the dam at the ith moment is obtained through calculationi Virtualization of。
As a preferable technical measure:
dam monitoring data comprises reservoir water level Zi Practice ofDam surface temperature Ti Practice ofTemperature H of dam bodyi Practice ofDam body displacement Di In fact；
Reservoir level Zi In factObtained by measuring with a water level gauge, and the error covariance is recorded as EZ；
Temperature T of dam surfacei Practice ofObtained by resistance thermometer measurement, and the error covariance is recorded as ET；
Temperature H of dam bodyi Practice ofObtained by distributed fiber measurement, and the error covariance is recorded as EH；
Dam body displacement Di Practice ofThe method is obtained by measurement of a hydrostatic level gauge and a dip angle multi-parameter sensor and a vertical line method, and the error covariance is recorded as ED；
Will include reservoir level Zi Practice ofDam surface temperature Ti In factTemperature H of dam bodyi In factDam body displacement Di In factAnd recombining the data according to spatial distribution to obtain a data file of the dam at the ith moment.
As a preferable technical measure:
in the third step, the data assimilation model is a cost function optimization parameter inversion model based on a three-dimensional variational method, and the construction method comprises the following steps: due to the nonlinear evolution characteristic of the concrete material along with time, the virtual displacement field T at the ith momenti Virtualization ofAnd the actual displacement field Di Practice ofIntroducing a data assimilation model to perform displacement inverse analysis on the concrete material line elastic constitutive parameters of the dam when the large difference exists;
taking initial material parameters of the dam as background values, setting a background error covariance matrix as a diagonal matrix of an initial value quadratic, taking the relative quantity of actual measuring point displacement of the dam at a certain time and a reference displacement initial value as an observed value, obtaining an observation error covariance matrix, and adjusting the observation error covariance matrix according to equipment errors;
meanwhile, thermal solid structure coupling simulation calculation of an initial physical field is used as a cost function of the data assimilation model, and the data assimilation model with optimized parameters is established;
according to the error covariance matrix of each dam monitoring data, a complete background error covariance matrix and an observation error covariance matrix are constructed for respectively expressing the reliability of background information and observation information in analysis, the reliability depends on the statistical characteristics of respective errors and the incidence relation among different variable errors, and the specific calculation formula is as follows:
wherein x isbAs a result of the background value,
b is the complete background error covariance matrix,
xi,0the value of the first guess is the initial guess value,
r is the complete observation error covariance matrix,
EZis reservoir level Zi In factThe error covariance matrix is then calculated,
ETfor the temperature T of the dam facei Practice ofThe error covariance matrix is then calculated,
EHtemperature of dam body Hi In factThe covariance matrix of the error is then determined,
EDfor dam body displacement Di In factAn error covariance matrix;
although improving the error estimation accuracy cannot ensure that the analysis value obtained each time is optimal, the probability of obtaining the optimal result can be significantly improved;
the cost function calculation formula is as follows:
wherein, yi oAs actual observed value, yi cAnd x is a concrete material line elastic constitutive parameter of the dam, which is an unknown but bounded interval variable.
As a preferable technical measure:
the calculation method of the elastic constitutive parameters of the concrete material lines comprises the following steps:
judging and giving an estimation range according to the concrete material test piece, and recording the estimation range as a feasible region DNThe specific expression is as follows：
In the formula aj、bjIs the upper and lower limit values of the jth concrete material parameter, x = { x = { (X)j}；
The search method of the nonlinear simplex method is used, the estimation range is adjusted,
constructing a sequence of one-step variable allowable tolerance functions φ0≥φ1≥…≥φkMore than or equal to 0}, the sequence monotonically decreases towards zero along with the iterative search times;
wherein phi iskIs a tolerance criterion function, which is a function of the positive simplex vertex, and the specific calculation formula is as follows:
wherein d is the simplex side length, xj kIs feasible domain DNThe jth vertex of the middle simple shape;
for the displacement inverse analysis problem, since the cost function is generally a positive number,
the cost function can thus be rewritten as a tolerance criterion function
And constraint violation estimator T (x)k) The inequality of (a):
while constraining the corruption estimator T (x)k) Indicating the extent to which the variables do not satisfy the constraints,
when T (x)k) Variable x when =0kAll the constraint conditions are satisfied;
when T (x)k) When not equal to 0, in the feasible region and the approximate feasible region, searching T (x) by gradually iteratingk)≤φkThe convergence speed then coincides with the decreasing speed of the variable tolerance function sequence.
As a preferable technical measure:
in the fourth step, the prediction model can simulate and predict the dam state from a certain moment to the next moment, and the construction method comprises the following steps:
step one, acquiring an assimilation temperature field T of a dam at the ith moment based on a data assimilation modeli Assimilation ofAnd assimilation of displacement field Di AssimilationAnd concrete line elastic constitutive parameter xAssimilation={xj AssimilationThe initial model physical field and the initial parameters of the prediction model of the dam are used as the parameters;
establishing a dam global physical state prediction model based on data assimilation by combining a finite element analysis method according to the initial model physical field and the initial parameters, and the environment temperature, the water temperature before the dam and the reservoir water level of the dam which are acquired in real time;
predicting the physical quantity of the dam, which evolves along with the time, from the ith moment to the (i + 1) th moment according to the dam global physical state prediction model;
the physical quantities comprise a temperature field, a displacement field, a stress field and a strain field;
and a fifth step of repeatedly executing the first step to the fourth step to obtain short-term global physical state information of the dam after each monitoring.
In order to achieve the above object, a second technical solution of the present invention is:
a dam state prediction system based on data assimilation,
applying the dam state prediction method based on data assimilation;
the system comprises a grid model generation module, a monitoring data processing module, a finite element calculation module, a data assimilation calculation module and a prediction analysis module;
the grid model generation module is used for establishing a three-dimensional structure hexahedral grid model of the dam based on each dam section, the water levels of the upstream and downstream of the dam body and the bedrock of the dam, setting linear units, secondary units and grid fineness;
the monitoring data processing module is used for reading, storing and preprocessing the monitoring data of the dam at each moment;
the finite element calculation module is used for acquiring a thermodynamic coupling model of a three-dimensional structure formed by boundary conditions, combining finite element simulation analysis, and taking a certain time as an initial time to obtain information of the change of a global physical field of the dam along with time within a short period from the certain time to the next time;
the data assimilation calculation module is used for data assimilation calculation of a dam displacement field, is provided with a three-dimensional variational method and a data assimilation model, adaptively constructs a cost function of the data assimilation model according to an actual measuring point and a virtual measuring point, and simultaneously provides an unconstrained and constrained implicit nonlinear optimization algorithm and a solver, accelerates the solving process of data assimilation and finally provides material assimilation parameters of a dam;
the prediction analysis module is used for predicting and analyzing physical field information of the dam, constructing a short-term dam global state prediction model from a certain moment to the next moment by using the acquired assimilation and monitoring data, outputting the temperature, displacement and stress conditions of any virtual measuring point in the dam space and time, providing output of an extreme value and a mean value, and having a function of storing prediction analysis data;
generating a three-dimensional structure hexahedral mesh model of the dam by using a mesh model generation module, acquiring original data or processed data required by finite element analysis and data assimilation calculation by combining a monitoring data processing module, carrying out thermosetting coupled dam finite element analysis by using the finite element calculation module to obtain an initial physical field of the dam, and then carrying out calculation by using a data assimilation model of the data assimilation calculation module; and in the period, calling a finite element calculation module to perform finite element analysis, obtaining a prediction analysis module of the dam formed by the physical field assimilated at a certain time, and operating the prediction analysis module to predict the short-term physical state of the dam in real time.
Through continuous exploration and test, the nonlinear evolution characteristic of the concrete material along with time is fully considered, namely the current situation that the virtual displacement field and the actual displacement field of the concrete material have larger difference along with the change of time, and the real-time physical state of the dam in service can be effectively predicted by constructing a grid model generation module, a monitoring data processing module, a finite element calculation module, a data assimilation calculation module and a prediction analysis module, so that the safety monitoring and real-time early warning of the dam are supported, and a basis is provided for daily operation and maintenance, natural disasters such as flood, sudden temperature rise and drop, strong snowfall and the like or reservoir water level adjustment under extreme weather.
Furthermore, compared with nonlinear calculation considering creep and hydration heat, the method effectively improves the calculation efficiency, can ensure better accuracy of results obtained through data assimilation calculation in short-term prediction, and can provide important reference data for real-time early warning analysis of the dam.
As a preferable technical measure:
monitoring data including environment quantity and deformation quantity, sequencing the data in a time and space mode to form an original data file for storage, and providing a preprocessing function;
preprocessing comprises smooth interpolation and abnormal point elimination of regression analysis;
the environmental quantities comprise upstream water level, downstream water level, air temperature, precipitation, dam water temperature and air pressure;
the deformation comprises dam body surface displacement, dam body internal displacement and inclination;
the boundary conditions include a dam mesh model and dam monitoring information.
Compared with the prior art, the invention has the following beneficial effects:
through continuous exploration and test, the nonlinear evolution characteristic of the concrete material along with time is fully considered, namely the current situation that the virtual displacement field and the actual displacement field of the concrete material have larger difference along with the change of time is considered, and the concrete material linear elastic constitutive parameters of the dam are subjected to displacement inverse analysis by constructing a data assimilation model; and the observation data is introduced into the data assimilation model to update the model, the influence of data errors on the model is fully considered, and the evolution direction of the data assimilation model is adjusted in real time, so that the estimation precision of the data assimilation model is improved, the prediction capability of the prediction model is effectively improved, and the scheme is simple, practical and feasible.
Furthermore, compared with nonlinear calculation considering creep and hydration heat, the calculation efficiency is effectively improved, and the result obtained through data assimilation calculation can ensure better accuracy in short-term prediction, so that important reference data can be provided for real-time early warning analysis of the dam.
Furthermore, the invention can obtain a prediction model of the dam formed by the physical field after assimilation at a certain moment, and can predict the short-term physical state of the dam in real time; therefore, the safety monitoring and real-time early warning of the dam are supported, and a basis is provided for daily operation and maintenance, natural disasters such as flood, sudden temperature rise and fall, strong snowfall and the like or reservoir water level adjustment in extreme weather.
Drawings
FIG. 1 is a schematic flow chart of a real-time dam state prediction method according to the present invention;
FIG. 2 is a schematic diagram of a dam model for finite element simulation analysis according to the present invention;
fig. 3 is a schematic structural diagram of a dam state real-time prediction system according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
On the contrary, the invention is intended to cover alternatives, modifications, equivalents and alternatives which may be included within the spirit and scope of the invention as defined by the appended claims. Furthermore, in the following detailed description of the present invention, certain specific details are set forth in order to provide a better understanding of the present invention. It will be apparent to one skilled in the art that the present invention may be practiced without these specific details.
As shown in fig. 1-2, an embodiment of the dam condition prediction method of the present invention:
a dam state prediction method based on data assimilation,
the method comprises the following steps:
the first step is as follows: obtaining the stress and deformation displacement of the dam as the initial physical field of the dam;
the second step: performing thermal coupling calculation according to the initial physical field in the first step to obtain a dam body virtual displacement field at a certain moment;
the third step: constructing a data assimilation model by using the dam body virtual displacement field in the second step and the displacement of the actual measuring point of the dam at the same time, and performing displacement inverse analysis on the linear elasticity constitutive parameters of the concrete material of the dam;
observation data are introduced into the data assimilation model to update the model, and the evolution direction of the data assimilation model is adjusted in real time to improve the estimation precision of the data assimilation model;
the fourth step: and calculating material parameters through the data assimilation model in the third step, and constructing a prediction model according to the collected dam monitoring data to predict the dam state.
Through continuous exploration and test, the nonlinear evolution characteristic of the concrete material along with time is fully considered, namely the current situation that the virtual displacement field and the actual displacement field of the concrete material have larger difference along with the change of time is fully considered, and the concrete material linear elasticity constitutive parameters of the dam are subjected to displacement inverse analysis by constructing a data assimilation model; and the observation data is introduced into the data assimilation model to update the model, the influence of data errors on the model is fully considered, and the evolution direction of the data assimilation model is adjusted in real time, so that the estimation precision of the data assimilation model is improved, the prediction capability of the prediction model is effectively improved, and the scheme is simple, practical and feasible.
The invention discloses a best embodiment of a dam state prediction method, which comprises the following steps:
a dam state prediction method based on data assimilation comprises the following steps:
s1: regular hexahedral dam grid units are constructed based on dam design parameters, the influence of the self weight of the structure and the dry shrinkage of concrete is considered, simulation calculation is carried out on the dam construction process, the horizontal stress among transverse seams of different dam sections is eliminated, and the stress and deformation displacement of the dam are obtained and used as the initial physical field of the dam.
S2: and taking the monitoring data of the dam at the current moment as an opening boundary condition, considering the environment temperature, the water temperature before the dam, the reservoir water level of the dam and the sediment deposition effect, and combining an initial physical field of the dam to perform thermodynamic coupling calculation to obtain a dam body virtual displacement field at the current moment.
S3: considering the initial material parameter of the dam as a background value, setting a background error covariance matrix as a diagonal matrix of an initial value quadratic, taking the relative quantity of the actual measuring point displacement of the dam at the current moment and the initial value of the reference displacement as an observed value, adjusting the observation error covariance matrix according to the equipment error, and establishing a mathematical model with optimized parameters by taking the thermal consolidation structure coupling simulation calculation of the initial physical field into consideration as a cost function of a data assimilation model.
S4: and (3) according to the optimal material parameters obtained by the data assimilation model and dam monitoring data collected in real time, including the ambient temperature, the water temperature before the dam, the reservoir water level and the like, simulating and predicting the dam state from the current moment to the next moment.
S5: and repeatedly executing the steps S2-S4 to obtain the short-term global physical state information of the dam after each monitoring update.
According to a concrete sampling test piece of the dam, elastic material parameters including elastic modulus, Poisson's ratio, thermal expansion coefficient, density, thermal conductivity and specific heat capacity are obtained. And simultaneously, combining the geometrical shape of the dam, constructing a three-dimensional structure hexahedral mesh model of the dam, and constructing a pouring model of each dam segment by using the three-dimensional structure model, namely the hydration heat reaction and the drying shrinkage process of concrete. In the concrete drying process, the stress release phenomenon exists at the interface between the dam segments, and displacement deformation and residual stress are formed.
Specifically, the number of each measurement time of the current dam is set to be i, i =0 is set to be the initial physical field measurement time of the dam, and the corresponding state is calculated as S1. Recording the dam material parameter at the moment when i =0 as a background value xb。
Assume that the current time is the time of the ith update of monitoring data.
The monitoring data of the dam at the current moment comprises reservoir water level Zi In fact: obtained by water level gauge measurement, and error covariance thereof is recorded as EZ(ii) a Temperature T of dam surfacei In factObtained by resistance thermometer measurement, the error covariance of which is denoted as ET(ii) a Temperature H of dam bodyi Practice ofObtained by distributed fiber measurement, and the error covariance is recorded as EH(ii) a Dam body displacement Di In factThe method is obtained by measuring a hydrostatic level and sensors with multiple parameters of inclination angle and the like and a vertical line method, and the error covariance is recorded as ED. And recombining the measured data according to spatial distribution to obtain a data file of the dam at the ith moment.
Dam facing temperature Ti Practice ofThe dam body temperature H is introduced by taking the autogenous heat source such as hydration heat reaction of the dam into consideration by taking hexahedral primary grid as a discrete unit as a heat source boundary of the dami In factAnd a correction value thetaiAnd calculating to obtain a virtual temperature field T of the dam at the ith momenti Virtualization of。
The virtual temperature field T at the ith momenti Virtualization ofSetting the water level Z before dam as the thermal boundaryi In factAs the pressure boundary of the dam, only considering the action of the heat boundary and the pressure boundary to carry out thermal coupling analysis, and calculating to obtain the virtual displacement field T of the dam at the ith momenti Virtualization of。
Due to the nonlinear evolution characteristic of the concrete material along with time, the virtual displacement field T at the ith momenti VirtualizationAnd the actual displacement field Di In factAnd introducing a data assimilation model to perform displacement inverse analysis on the concrete material line elastic constitutive of the dam when a large difference exists. Setting the background value x as previously describedbThe background error covariance matrix B. Setting the initial value as the initial guess value xi,0And considering the error covariance matrix of each monitoring data to construct a complete background error covariance matrix B and an observation error covariance matrix R:
the covariance of the background and the observation errors is used for expressing the reliability of the background information and the observation information in the analysis, and generally depends on the statistical characteristics of respective errors and the incidence relation between different variable errors, and although the improvement of the error estimation precision cannot ensure that the analysis value obtained each time is optimal, the probability of obtaining the optimal result can be obviously improved.
Setting an actual observed value yi oIs the actual displacement field D of the dami，
Virtual observed value yi cVirtual displacement field D for dami VirtualizationAnd constructing a data assimilation model of the dam, namely a cost function optimization parameter inversion model based on a three-dimensional variational method, wherein a cost function J is as follows:
wherein x is the elastic constitutive parameter of the concrete material line of the dam and is an unknown but bounded interval variable. The estimation range is judged and given according to the concrete material test piece and is recorded as a feasible region DNNamely:
in the formula aj、bjX = { x) for upper and lower limit values of jth concrete material parameterj}. Using a search method of a nonlinear simplex method to adjust the estimation range and construct a one-step variable allowable tolerance function sequence (phi)0≥φ1≥…≥φk≧ 0}, the sequence monotonically decreases toward zero with the number of iterative searches.Function of tolerance criterion phikIs a function of the simplex vertex for positive values:
wherein d is the simplex side length, xj kIs feasible domain DNThe jth vertex of the middle simple shape.
For the displacement inverse analysis problem, since the cost function is generally positive, the cost function can be rewritten as a tolerance criterion function φkAnd constraint violation estimator
The inequality of (a):
while constraining the corruption estimator T (x)k) Indicating the degree to which the variable does not satisfy the constraint, when T (x)k) When =0, variable xkAll the constraint conditions are satisfied; when T (x)k) When not equal to 0, in the feasible region and the approximate feasible region, searching T (x) by gradually iteratingk)≤φkThe convergence speed then coincides with the decreasing speed of the variable tolerance function sequence.
Acquiring an assimilation temperature field T of the dam at the ith moment according to the data assimilation modeli AssimilationAnd assimilation of displacement field Di Transforming into And concrete linear elastic constitutive parameter xAssimilation={xj Assimilation ofAnd (4) as the initial physical field and initial parameters of the prediction model of the dam.
And establishing a real-time prediction model by using the dam state information acquired in real time. Specifically, a data assimilation-based dam global physical state prediction model is established by utilizing real-time monitoring information of the ith moment, including dam front water temperature, air temperature and upstream and downstream water levels, and combining a finite element analysis method, and the model is used for predicting physical fields such as a temperature field, a displacement field, a stress field and a strain field of the dam from the ith moment to the (i + 1) th moment, which evolve along with time.
The prediction method is based on the concrete line elastic constitutive of the dam, the calculation efficiency is effectively improved in comparison with the nonlinear calculation considering creep and hydration heat, the result obtained through data assimilation calculation can ensure better accuracy in short-term prediction, and important reference data can be provided for real-time early warning analysis of the dam.
As shown in fig. 3, a specific embodiment of the dam deformation prediction system based on data assimilation of the present invention:
a dam deformation prediction system based on data assimilation, comprising:
the grid model generating module 10 is used for establishing a three-dimensional structure hexahedral grid model of the dam based on each dam section, the water levels of the upstream and downstream of the dam body and the bedrock of the dam, setting linear units, secondary units and grid fineness;
and the monitoring data processing module 20 is used for reading, storing and preprocessing the monitoring information of the dam at each moment. The monitoring data comprises environmental quantities such as upstream and downstream water levels, air temperature, precipitation, dam water temperature and air pressure and deformation quantities such as dam body surface displacement, dam body internal displacement and inclination, the environmental quantities are sequenced in a time and space mode to form an original data file for storage, and preprocessing functions such as smooth interpolation, abnormal point elimination of regression analysis and the like are provided;
the finite element calculation module 30 is used for forming a thermal coupling model of a three-dimensional structure by using the acquired boundary conditions such as the dam grid model and the dam monitoring information, combining finite element simulation analysis, and taking the current moment as an initial moment to obtain information of the change of a global physical field of the dam along with time in a short period from the current moment to the next moment;
the data assimilation calculation module 40 is used for data assimilation calculation of a dam displacement field, is provided with a data assimilation model such as a three-dimensional variational method and the like, adaptively constructs a cost function of the data assimilation model according to an actual measuring point and a virtual measuring point, and simultaneously provides an unconstrained and constrained implicit nonlinear optimization algorithm and a solver, accelerates the solving process of data assimilation, and finally provides material assimilation parameters of the dam;
and the prediction analysis module 50 is used for predicting and analyzing the physical field information of the dam, constructing a short-term dam global state prediction model from the current moment to the next moment by using the acquired assimilation and monitoring data, outputting the temperature, displacement and stress conditions of any virtual measuring point in the dam space and time, outputting an extreme value and a mean value, and storing prediction analysis data.
It can be seen from the foregoing technical solutions that, the present embodiment provides a dam state real-time prediction system based on data assimilation, which specifically includes: the three-dimensional structure hexahedral mesh model of the dam is generated by using the mesh model generation module 10, original data or processed data required by finite element analysis and data assimilation calculation are acquired by combining the monitoring data processing module 20, the dam finite element analysis of thermosetting coupling is carried out by using the finite element calculation module 30 to obtain an initial physical field of the dam, and then calculation is carried out by using the data assimilation model of the data assimilation calculation module 40. And in the period, calling the finite element calculation module 30 to perform finite element analysis to obtain a prediction analysis module of the dam formed by the physical field assimilated at the current moment, and operating the prediction analysis module 50 to predict the short-term physical state of the dam in real time. Through the processing, the real-time physical state of the dam in service can be effectively predicted, so that the safety monitoring and real-time early warning of the dam are supported, and a basis is provided for daily operation and maintenance, and reservoir water level adjustment under natural disasters such as flood, sudden temperature rise and drop, strong snow fall and the like or extreme weather.
One specific embodiment of the application of the invention:
in a certain concrete hyperbolic high arch dam project in the middle of China, the height of a dam body is 103 meters, the arc length of a dam top is 227.9m, the rainwater collection area is 266.1 square kilometers, the total storage capacity is 8257 ten thousand cubic meters, and a power station is installed with 6 ten thousand kilowatts, so that the dam is one of important clean power sources in the local. The arch dam project is started in 4 months in 2002 and finished at the end of 9 months in 2005. The dam body is made of C20W8F50 concrete, the total pouring concrete is 210458 cubic meters, 13 dam sections are irrigated in two orders, the dam body is relatively thin and is easily affected by water level and weather, and therefore the arch dam needs to be monitored in real time to prevent accidents of the arch dam.
The arch dam has multiple monitoring modes, including but not limited to deformation monitoring, leakage amount monitoring, dam foundation uplift pressure monitoring, dam-surrounding seepage monitoring, stress strain monitoring, bedrock deflection monitoring, seam monitoring, dam body temperature monitoring, air temperature monitoring, water level monitoring and the like, and 3 horizontal observation sections (442 meters, 397 meters and 375 meters in height) and 3 vertical observation sections (No. 5, No. 8 and No. 11 dam sections) are arranged.
Firstly, a grid model generating module is used for generating a grid model of the arch dam, and the method comprises the following specific steps:
1. according to the physical information and design parameters of the arch dam, drawing a geometric model of the arch dam by using a CAD modeling tool: the whole body is a regular hexahedron, but because the dam site valley is a V-shaped canyon with two sides not aligned, the reserved widths on two sides of the geometric model are unequal, and meanwhile, the edge effect calculated by finite elements is considered, and the length of the dam height of a bedrock part below the riverbed is prolonged by about 1.5 times.
2. And partitioning the CAD model according to the pouring process of the arch dam, so that the hexahedral mesh can be generated in a self-adaptive manner.
3. And setting grid division parameters according to primary and secondary objects of finite element analysis, wherein the density of the bedrock grid is rough, and the density of the dam body grid is relatively fine. And completing grid sensitivity analysis in the primary calculation process to obtain the optimal grid division parameters.
4. According to the method, the hexahedral mesh of the arch dam is generated.
According to the basic principle of data assimilation, the initial physical field of the arch dam needs to be obtained as the initial condition of assimilation calculation, so that the completion state of the arch dam is calculated as the initial physical field, and the method specifically comprises the following steps:
1. and dividing the 13 dam segments into odd-even segments for calculation. The odd sections are dam sections for the first-order irrigation, and the even sections are dam sections for the second-order irrigation.
2. Acquiring pure elastic parameters and concrete drying shrinkage parameters of different dam sections at each time period according to the physical property parameter curve of the concrete sample in the arch dam 50-month construction process: young's modulus, Poisson's ratio, coefficient of thermal expansion, thermal conductivity, volume specific heat capacity, coefficient of dry shrinkage, coefficient of autogenous shrinkage, and water content.
3. Considering the self-weight of the structure, calculating the physical field of the odd dam sections before the second-order irrigation according to the concrete hydration thermal model, after eliminating the horizontal stress of the transverse seam sections, calculating the even dam sections, and acquiring the stress and displacement of the arch dam during the completion as the initial physical field S of the damInitiation ofAnd DInitialSimultaneously recording the material parameter value of each dam section of the arch dam as a background value xb。
Acquiring each open boundary parameter and a measurement error covariance matrix at the ith moment by using a monitoring tool and a monitoring data processing function of a monitoring data processing module, wherein the parameters are as follows:
1. the vertical displacement monitoring is obtained by a level gauge with the same precision of 'NA 2+ GPM', and the measurement precision is +/-1.0 mm;
2. the horizontal displacement observation is obtained by a total station instrument with the same precision of 'come card TCA 1800', and the measurement precision is +/-2.0 mm;
3. temperature monitoring is obtained through a resistance thermometer, the measurement precision is +/-1 ℃, and buried points are uniformly paved inside the dam body and on the surface of the dam body;
4. reservoir water level monitoring is obtained through a vertical electronic water gauge, and the measurement precision is +/-1 cm;
5. normalizing the measurement precision of the displacement, temperature and water level monitoring data to obtain an error covariance matrix of each monitoring data: eZ、ET、EHAnd ED…, it is noted that the products of all normalized covariance and measured data should be in the same order of magnitude to ensure the convergence trend of each variable to be consistent, so that the obtained data assimilation result is more reliable;
6. and recombining the measured data according to the spatial distribution of the arch dam to obtain an ith moment data file of the dam, and storing the ith moment data file into a universal comma separated value csv file format.
Reading csv format file, taking the monitoring data of the arch dam at the i-th moment, and calling API (application program interface) input of a finite element computing module to serve as the interface inputThe thermal boundary condition of the arch dam is specifically that a hexahedral primary grid is taken as a discrete unit, autogenous heat sources such as hydration heat reaction and the like are considered, the temperature and the corrected value of the dam body are introduced, and the virtual temperature field T of the dam at the ith moment is calculated and obtainedi Virtualization of. Wherein, the correction value is set as the unbiased estimation variance of each actual measuring point and each virtual measuring point.
And according to the precision requirement of mechanical calculation, calculating the thermal stress and the mechanical stress by using a quadratic grid. Specifically, the virtual temperature field T at the ith moment is determinedi Virtualization ofThe hexahedral secondary grid mapped to the dam is used as a thermal boundary, the water level in front of the dam is used as a pressure boundary of the dam, thermal coupling analysis is carried out only by considering temperature and mechanical action, and a virtual displacement field D of the dam at the ith moment is obtained through calculationi Virtualization of。
At the moment i, the thermodynamic coupling calculation of the arch dam is completed, but the nonlinear evolution characteristic of the concrete material along with the time Di VirtualizationAnd the actual measured displacement Di Practice ofAnd (4) carrying out displacement inverse analysis based on data assimilation calculation on the concrete material line elastic constitutive of the dam by using a data assimilation calculation module when a large difference exists. The method comprises the following specific steps:
1. setting the background value x as described previouslybThe background error covariance matrix B. The NMC method is selected as a background error covariance simulation method, namely interpolation between two predicted values of different predicted aging at the same moment is used as a background error, and the predicted values about 15 days are obtained according to experience:
in the formula xi tThe predicted value of 15 days at the same time;
2. setting an actual observed value yi oIs the actual displacement field D of the dami Practice ofCombining the measurement error covariance matrixes to construct an overall measurement error covariance matrix R:
when the number of the measuring points of the arch dam is 38, the R matrix is a 38 x 38 matrix;
3. virtual observed value yi cVirtual displacement field D for dami VirtualizationAnd constructing a data assimilation model of the dam, namely taking a cost function J based on a three-dimensional variational method as an optimization parameter inversion model. Note that the process of computing the cost function J necessarily introduces a virtual measurement point yi cAnd obtaining the data through finite element simulation calculation. Namely, the finite element calculation module is required to be called to calculate the cost function J in each step;
4. the limited estimation range is reasonably given according to the concrete material test piece and is recorded as a feasible region DNNamely:
in the formula aj、bjX = { x) for upper and lower limit values of jth concrete material parameterj}。
5. Constructing a sequence of tolerance criteria functions φ0≥φ1≥…≥φk≧ 0}, the sequence monotonically decreases with the number of iterative searches toward zero as a function of its positive simplex vertex.
6. Setting D as feasible region DNLength of side of simplex, xj kIs feasible domain DNThe jth vertex of the middle simple shape. Introducing and solving a constraint violation estimator T (x)k) Phi of inequalityk-T(xk) More than or equal to 0, the steps are as follows:
a) giving an initial value xbAnd side length d from xbStarting to perform unconstrained simplex acceleration method descent iteration on J (x);
b) with xbCalculating each simplex vertex x for the center point and d for the side length1 b、x2 b、…、xN+1 bValue of cost function ofJ(xi b)，
i is 1, 2, …, N +1, then finding the optimal point xl bAnd worst point xh bAnd calculating the central points of all the points except the worst point;
c) calculate the optimal point xl bT (x) ofl b) Checking phi0-T(xl b) Whether or not 0 or more holds:
i. if true, xl bIn feasible domain DNIn the inner or approximate feasible region, a new point can be obtained by a simplex method and is used for replacing the worst point;
if not, minimizing T (x) by simplex accelerationl b) Solving a point to replace the new point, wherein the new point meets the inequality;
d) the calculation is converted from a minimization search of t (x) back to a minimization search of j (x).
In a new round of search, let k equal to k +1, the program gives the convergence criterion epsilon and discriminates phikWhether or not < ε is true:
i. if yes, outputting a calculation result and stopping the minimization search;
if not, a new round of cost function J (x) search is started.
At the moment, the assimilation temperature field T of the dam at the ith moment is obtainedi Assimilation ofAnd assimilation of the displacement field Di Assimilation ofAnd concrete line elastic constitutive parameter xi AssimilationAnd calling a prediction analysis module to predict the real-time state of the arch dam, specifically, establishing a data assimilation-based dam global physical state prediction model by using real-time monitoring information at the ith moment, including the water temperature, air temperature and upstream and downstream water levels, in combination with a finite element analysis method, and predicting the physical fields of the dam, such as a temperature field, a displacement field, a stress field, a strain field and the like, which evolve along with time from the ith moment to the (i + 1) th moment.
After the steps, a 3D cloud picture of the physical field of the arch dam at each moment can be seen on a computer, and the maximum value, the minimum value and the global mean value of important variables such as displacement, temperature, stress and the like and the positions of the important variables on the 3D cloud picture are marked on the cloud picture. Meanwhile, the prediction analysis module can support the physical quantity prediction of each virtual point through Gaussian smooth interpolation calculation, and can provide a high-precision and practical reference basis for the real-time safety of the operation and maintenance of the arch dam by combining the data analysis of the monitoring system. In addition, prediction of extreme weather conditions is supported, and compared with the traditional method, the method can be used for estimating the dam global state under the extreme weather conditions more quickly and accurately.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (9)
1. A dam state prediction method based on data assimilation is characterized in that,
the method comprises the following steps:
the first step is as follows: obtaining the stress and deformation displacement of the dam as the initial physical field of the dam;
the second step is that: performing thermal coupling calculation according to the initial physical field in the first step to obtain a dam body virtual displacement field at a certain moment;
the third step: constructing a data assimilation model by using the dam body virtual displacement field in the second step and the displacement of the actual measuring point of the dam at the same time, and performing displacement inverse analysis on the linear elasticity constitutive parameters of the concrete material of the dam;
the data assimilation model is a cost function optimization parameter inversion model based on a three-dimensional variational method, and the construction method comprises the following steps:
taking initial material parameters of the dam as background values, setting a background error covariance matrix as a diagonal matrix of an initial value quadratic, taking the relative quantity of actual measuring point displacement of the dam at a certain moment and a reference displacement initial value as an observed value to obtain an observed error covariance matrix, and adjusting the observed error covariance matrix according to equipment errors;
meanwhile, thermal solid structure coupling simulation calculation of an initial physical field is used as a cost function of the data assimilation model, and the data assimilation model with optimized parameters is established;
according to the error covariance matrix of each dam monitoring data, a complete background error covariance matrix and an observation error covariance matrix are constructed for respectively expressing the reliability of background information and observation information in analysis, the reliability depends on the statistical characteristics of respective errors and the incidence relation among different variable errors, and the specific calculation formula is as follows:
wherein B is a complete background error covariance matrix,
xbin the case of the background value, the value,
xi,0the value of the first guess is the initial guess value,
r is the complete observation error covariance matrix,
EZis the front water level Z of the dami In factThe error covariance matrix is then calculated,
ETfor the temperature T of the dam facei In factThe error covariance matrix is then calculated,
EHis the temperature H of the dam bodyi In factThe covariance matrix of the error is then determined,
EDfor dam body displacement Di In factAn error covariance matrix;
the cost function calculation formula is as follows:
wherein, yi oAs a real observed value, yi cThe method is characterized in that the method is a virtual observation value, x is a concrete material line elastic constitutive parameter of a dam and is an unknown but bounded interval variable;
observation data are introduced into the data assimilation model to update the model, and the evolution direction of the data assimilation model is adjusted in real time according to the influence of data errors on the model;
the fourth step: and calculating material parameters through the data assimilation model in the third step, and constructing a prediction model according to the collected dam monitoring data to predict the dam state.
2. The dam state prediction method based on data assimilation of claim 1, characterized in that,
in the first step, the method for acquiring the stress and the deformation displacement comprises the following steps:
constructing a regular three-dimensional structure hexahedral mesh model based on the design parameters of the dam and the parameters of the elastic material, and establishing a pouring model of each dam segment according to the three-dimensional structure hexahedral mesh model, wherein the pouring model is used for reflecting the hydration heat reaction and the drying shrinkage process of concrete; and according to the pouring model, carrying out simulation calculation on the dam construction process to obtain the stress and deformation displacement of the dam.
3. The dam state prediction method based on data assimilation of claim 2,
the elastic material parameters comprise elastic modulus, Poisson's ratio, thermal expansion coefficient, density, thermal conductivity and specific heat capacity, and are obtained by detecting a concrete sampling test piece of the dam.
4. The dam state prediction method based on data assimilation of claim 1,
in the second step, the method for acquiring the virtual displacement field is as follows:
dam surface temperature T is set according to ambient temperature, water temperature before dam, dam reservoir water level and silt deposition effecti Practice ofAs the heat source boundary of the dam, a hexahedral primary grid is specifically taken as a discrete unit;
introducing the dam body temperature H according to the hydration heat reaction and the self-generated heat source existing in the dami Practice ofAnd a correction value thetaiAnd calculating to obtain the virtual temperature field T of the dam at the ith momenti Virtualization of；
The virtual temperature field T at the ith momenti Virtualization ofThe hexahedral quadratic mesh mapped to the dam serves as a thermal boundary,
setting the front water level Z of the dami Practice ofAs the pressure boundary of the dam, only the action of the heat boundary and the pressure boundary is utilized to carry out thermal coupling analysis, and the virtual displacement field T of the dam at the ith moment is obtained through calculationi Virtualization。
5. The dam state prediction method based on data assimilation of claim 4,
the dam monitoring data comprises dam front water level Zi In factDam surface temperature Ti Practice ofTemperature H of dam bodyi Practice ofDam body displacement Di In fact：
Water level Z in front of dami Practice ofObtained by measuring with a water level gauge, and the covariance of the error is recorded as EZ；
Temperature T of dam surfacei In factObtained by resistance thermometer measurement, the error covariance of which is denoted as ET；
Temperature H of dam bodyi Practice ofObtained by distributed fiber measurement, and the error covariance is recorded as EH；
Dam body displacement Di In factObtained by measurement of a hydrostatic level and a dip angle multi-parameter sensor and a vertical line method, and the error covariance is recorded as ED；
Will include the front water level Z of the dami Practice ofDam surface temperature Ti Practice ofTemperature H of dam bodyi Practice ofDam body displacement Di In factAnd recombining the data according to spatial distribution to obtain a data file of the dam at the ith moment.
6. The dam state prediction method based on data assimilation of claim 5,
the calculation method of the elastic constitutive parameters of the concrete material lines comprises the following steps:
judging and giving an estimation range according to the concrete material test piece, and recording the estimation range as a feasible region DNThe specific expression is as follows:
in the formula aj、bjIs the upper and lower limit values of the jth concrete material parameter, x = { x = { (X)j}；
The search method of the nonlinear simplex method is used, the estimation range is adjusted,
constructing a sequence of one-step variable allowable tolerance functions φ0≥φ1≥…≥φkMore than or equal to 0}, the sequence monotonically decreases towards zero along with the iterative search times;
wherein phi iskIs a tolerance criterion function, which is a function of the positive simplex vertex, and the specific calculation formula is as follows:
wherein d is the simplex side length, xj kIs feasible domain DNThe jth vertex of the middle simple body;
rewriting a cost function to a tolerance criterion function phikAnd constraint violation estimator T (x)k) The inequality of (a):
while constraining the corruption estimator T (x)k) Indicating the extent to which the variables do not satisfy the constraints,
when T (x)k) Variable x when =0kAll the constraint conditions are satisfied;
when T (x)k) When not equal to 0, in the feasible region and the approximate feasible region, searching T (x) by step-by-step iterationk)≤φkThe convergence speed then coincides with the decreasing speed of the variable tolerance function sequence.
7. The dam condition prediction method based on data assimilation of any one of claims 1 to 6,
in the fourth step, the prediction model can simulate and predict the dam state from a certain moment to the next moment, and the construction method comprises the following steps:
step one, acquiring an assimilation temperature field T of a dam at the ith moment based on a data assimilation modeli Assimilation ofAnd assimilation siteMoving field Di Transforming into And concrete linear elastic constitutive parameter xAssimilation of={xj AssimilationThe initial model physical field and the initial parameters of the prediction model of the dam are used as the parameters;
establishing a dam global physical state prediction model based on data assimilation by combining a finite element analysis method according to the initial model physical field and the initial parameters, and the environment temperature, the water temperature before the dam and the reservoir water level of the dam which are acquired in real time;
predicting the physical quantity of the dam, which evolves along with time from the ith moment to the (i + 1) th moment, according to the dam global physical state prediction model;
the physical quantities comprise a temperature field, a displacement field, a stress field and a strain field;
and a fifth step of repeatedly executing the first step to the fourth step to obtain short-term global physical state information of the dam after each monitoring.
8. A dam state prediction system based on data assimilation is characterized in that,
applying a dam condition prediction method based on data assimilation as claimed in any one of claims 1-7;
the system comprises a grid model generation module, a monitoring data processing module, a finite element calculation module, a data assimilation calculation module and a prediction analysis module;
the grid model generation module is used for establishing a three-dimensional structure hexahedral grid model of the dam based on each dam section, the water levels of the upstream and downstream of the dam body and the bedrock of the dam, setting linear units, secondary units and grid fineness;
the monitoring data processing module is used for reading, storing and preprocessing the monitoring data of the dam at each moment;
the finite element calculation module is used for acquiring a thermodynamic coupling model of a three-dimensional structure formed by boundary conditions, combining finite element simulation analysis, and taking a certain time as an initial time to obtain information of the change of a global physical field of the dam along with time within a short period from the certain time to the next time;
the data assimilation calculation module is used for data assimilation calculation of a dam displacement field, is provided with a three-dimensional variational method and a data assimilation model, adaptively constructs a cost function of the data assimilation model according to an actual measuring point and a virtual measuring point, and simultaneously provides an unconstrained and constrained implicit nonlinear optimization algorithm and a solver, accelerates the solving process of data assimilation, and finally provides material assimilation parameters of the dam;
the prediction analysis module is used for predicting and analyzing physical field information of the dam, constructing a short-term dam global state prediction model from a certain moment to the next moment by using the acquired assimilation and monitoring data, outputting the temperature, displacement and stress conditions of any virtual measuring point in the dam space and time, providing output of an extreme value and a mean value, and having a storage function of prediction analysis data;
generating a three-dimensional structure hexahedral mesh model of the dam by using a mesh model generation module, acquiring original data or processed data required by finite element analysis and data assimilation calculation by combining a monitoring data processing module, carrying out thermosetting coupled dam finite element analysis by using the finite element calculation module to obtain an initial physical field of the dam, and then carrying out calculation by using a data assimilation model of the data assimilation calculation module; and in the period, calling a finite element calculation module to perform finite element analysis, obtaining a prediction analysis module of the dam formed by the physical field assimilated at a certain time, and operating the prediction analysis module to predict the short-term physical state of the dam in real time.
9. The dam condition prediction system based on data assimilation of claim 8, wherein,
monitoring data comprises environment quantity and deformation, sequencing the environment quantity and the deformation in a time and space mode to form an original data file for storage, and providing a preprocessing function;
preprocessing comprises smooth interpolation and abnormal point elimination of regression analysis;
the environmental quantity comprises an upstream water level, a downstream water level, an air temperature, a precipitation, a dam water temperature and an air pressure;
the deformation comprises dam body surface displacement, dam body internal displacement and inclination;
the boundary conditions include a dam mesh model and dam monitoring information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210432633.4A CN114547951B (en) | 2022-04-24 | 2022-04-24 | Dam state prediction method and system based on data assimilation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210432633.4A CN114547951B (en) | 2022-04-24 | 2022-04-24 | Dam state prediction method and system based on data assimilation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114547951A CN114547951A (en) | 2022-05-27 |
| CN114547951B true CN114547951B (en) | 2022-07-22 |
Family
ID=81666947
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210432633.4A Active CN114547951B (en) | 2022-04-24 | 2022-04-24 | Dam state prediction method and system based on data assimilation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114547951B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115146565B (en) * | 2022-09-02 | 2022-11-25 | 浙江远算科技有限公司 | Hyperbolic arch dam foundation seepage evaluation method and system based on cohesion algorithm |
| CN116776644B (en) * | 2023-08-18 | 2023-10-27 | 浙江远算科技有限公司 | Dam operation monitoring simulation method and system based on digital twin |
| CN117272872B (en) * | 2023-11-21 | 2024-01-30 | 四川大学 | Panel rock-fill dam deformation monitoring method based on component separation |
| CN117690093A (en) * | 2024-01-31 | 2024-03-12 | 华能澜沧江水电股份有限公司 | Dam safety monitoring operation maintenance method and device, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009052976A (en) * | 2007-08-24 | 2009-03-12 | Toshiba Corp | Weather prediction data analyzer and weather prediction data analysis method |
| CN108334660A (en) * | 2017-12-28 | 2018-07-27 | 中国水利水电科学研究院 | A kind of the water sand prediction technique and system of the strong alluvial stream based on data assimilation |
| CN109992900A (en) * | 2019-04-08 | 2019-07-09 | 清华大学 | A kind of more real-time online cooperative intelligent emulation modes of mass concrete and system |
| CN110457651A (en) * | 2019-07-05 | 2019-11-15 | 中国地质大学（武汉） | It is a kind of to utilize the improved Prediction of Displacement in Landslide method of data assimilation |
| CN111259590A (en) * | 2020-01-17 | 2020-06-09 | 南昌大学 | Construction method of concrete dam deformation safety monitoring model |
| CN111414658A (en) * | 2020-03-17 | 2020-07-14 | 宜春学院 | Rock mass mechanics parameter inverse analysis method |
| CN111914496A (en) * | 2020-08-15 | 2020-11-10 | 四川大学 | Frozen reservoir water temperature-ice condition inversion and prediction method based on thermal coupling model |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10203268B2 (en) * | 2008-12-04 | 2019-02-12 | Laura P. Solliday | Methods for measuring and modeling the process of prestressing concrete during tensioning/detensioning based on electronic distance measurements |
| CN111695177B (en) * | 2020-05-14 | 2022-10-14 | 河海大学 | Dam mechanics parameter random inversion method and system based on displacement field monitoring data |
-
2022
- 2022-04-24 CN CN202210432633.4A patent/CN114547951B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009052976A (en) * | 2007-08-24 | 2009-03-12 | Toshiba Corp | Weather prediction data analyzer and weather prediction data analysis method |
| CN108334660A (en) * | 2017-12-28 | 2018-07-27 | 中国水利水电科学研究院 | A kind of the water sand prediction technique and system of the strong alluvial stream based on data assimilation |
| CN109992900A (en) * | 2019-04-08 | 2019-07-09 | 清华大学 | A kind of more real-time online cooperative intelligent emulation modes of mass concrete and system |
| CN110457651A (en) * | 2019-07-05 | 2019-11-15 | 中国地质大学（武汉） | It is a kind of to utilize the improved Prediction of Displacement in Landslide method of data assimilation |
| CN111259590A (en) * | 2020-01-17 | 2020-06-09 | 南昌大学 | Construction method of concrete dam deformation safety monitoring model |
| CN111414658A (en) * | 2020-03-17 | 2020-07-14 | 宜春学院 | Rock mass mechanics parameter inverse analysis method |
| CN111914496A (en) * | 2020-08-15 | 2020-11-10 | 四川大学 | Frozen reservoir water temperature-ice condition inversion and prediction method based on thermal coupling model |
Non-Patent Citations (6)
| Title |
|---|
| An Improved Random Forest Model for the Prediction of Dam Displacement;YAN SU,et al.;《IEEE》;20210106;第9142-9153页 * |
| Data assimilation (4D-VAR) to forecast flood in shallow-waters with sediment erosion;Eric Belanger,et al.;《Journal of Hydrology》;20040728;第114-125页 * |
| 不同机器学习模型对混凝土坝位移预测性能的影响;隋旭鹏 等;《水电能源科学》;20220131;第40卷(第1期);第107-111页 * |
| 基于Code_aster的拱坝数字孪生健康评估平台开发;罗增浤 等;《第十七届中国CAE工程分析技术年会论文集》;20211130;第246-253页 * |
| 基于有限元法的大坝应力应变计算及监测结果分析-基于NEDSK工程;王燕芬;《中国优秀硕士学位论文全文数据库 工程科技II辑》;20210415(第4期);C037-61 * |
| 大体积混凝土浇筑过程热应力分析和参数化模板开发;刘浩;《中国优秀博硕士学位论文全文数据库（硕士） 工程科技II辑》;20131215(第S2期);C038-596 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114547951A (en) | 2022-05-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114547951B (en) | Dam state prediction method and system based on data assimilation | |
| Jin et al. | Bayesian model selection for sand with generalization ability evaluation | |
| Su et al. | Time-varying identification model for dam behavior considering structural reinforcement | |
| CN106529055A (en) | Model updating method based on strain modal shape correlation | |
| CN103324798B (en) | Based on the stochastic response of interval response surface model | |
| CN104200005A (en) | Bridge damage identification method based on neural network | |
| CN112966856A (en) | Mountain torrent risk prediction method and prediction system | |
| CN107729592A (en) | Traced back the Time variable structure Modal Parameters Identification of track based on broad sense subspace | |
| CN103198215A (en) | Method for predicting subsidence of soil outside pit based on differential evolution support vector machine | |
| CN111695177A (en) | Dam mechanics parameter random inversion method and system based on displacement field monitoring data | |
| CN107657083A (en) | A kind of intellectual water meter is in Line synthesis error calibration method | |
| CN115935834A (en) | History fitting method based on deep autoregressive network and continuous learning strategy | |
| CN112307536B (en) | Dam seepage parameter inversion method | |
| CN117113854B (en) | Salt tide forecasting method based on ConvLSTM and three-dimensional numerical simulation | |
| CN112100711B (en) | ARIMA and PSO-ELM based concrete dam deformation combined forecasting model construction method | |
| CN110263469A (en) | A kind of improvement truncation Matrix Singular Value for overcoming structural model to correct pathosis | |
| CN112836789A (en) | Ground connection wall deformation dynamic prediction method based on composite neural network algorithm | |
| CN111859736A (en) | Service performance evaluation method for pipeline under multi-disease complex environment | |
| Li et al. | An ANN-based short-term temperature forecast model for mass concrete cooling control | |
| CN115455793A (en) | High-rise structure complex component stress analysis method based on multi-scale model correction | |
| Xu et al. | Improved hybrid model for predicting concrete crack openings based on chaos theory | |
| CN107908824A (en) | Reinforced concrete simple-supported beam fire model modification method based on power or force model | |
| CN114861489B (en) | rPCK proxy model aided structure dynamic parameter identification method | |
| Song et al. | Long-term deformation safety evaluation method of concrete dams based on the time-varying stability of concrete material | |
| Yang et al. | The influence of different back analysis conditions on the stochastic back analysis of arch dam material Parameters: A case study |
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 |