CN103353295A - Method for accurately predicating vertical deformation of dam body - Google Patents
Method for accurately predicating vertical deformation of dam body Download PDFInfo
- Publication number
- CN103353295A CN103353295A CN2013101891390A CN201310189139A CN103353295A CN 103353295 A CN103353295 A CN 103353295A CN 2013101891390 A CN2013101891390 A CN 2013101891390A CN 201310189139 A CN201310189139 A CN 201310189139A CN 103353295 A CN103353295 A CN 103353295A
- Authority
- CN
- China
- Prior art keywords
- model
- dam
- vertical deformation
- formula
- dam body
- 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
Abstract
The invention discloses a method for accurately predicating vertical deformation of a dam body. The method for accurately predicating the vertical deformation of the dam body is based on the stability. Specifically, the method comprises the steps of 1) collecting monitoring data; 2) building mathematical models of i kinds of predictions, wherein i is an integer which is greater than or equal to 2; 3) calculating the stability of various prediction methods; 4) determining a weight coefficient of a combined method model; and 5) building a stability based dam body vertical deformation predicting model. The prediction accuracy of the vertical deformation of the dam body can be greatly improved by adopting the method disclosed by the invention. Analysis of application results of a large number of engineering projects indicates that the prediction accuracy of the deformation is improved by 20%-70% by adopting the method disclosed by the invention compared with other prediction methods. The method can maintain the continuity of the model precision, and makes accurate predictions on the development trend of the safety related deformation, thereby being conducive to taking measures in advance, and having important significance for preventing safety accidents of a dam. The method disclosed by the invention has obvious social and economic values.
Description
Technical field
The invention belongs to the geodesy technical field in the Surveying Science and Technology subject, relate in particular to the Forecasting Methodology of dam vertical deformation amount.
Background technology
Along with the exploitation of waterpower resourses, the scale of dam is increasing, and the geologic condition of dam site also becomes increasingly complex.Therefore, the safety problem of dam merits attention.According to the whole nation for the third time statistics show, end 2003, China's dam-break accident has 3481.So, after dam builds up, in order to understand the running status of dam, ensure the safe operation of dam, all must carry out safety monitoring.The generation of dam accident is not accidental, generally all has progressively evolution from quantitative change to qualitative change.By analysis and the processing to the dam deformation data, the real-time working that can in time grasp dam is dynamic, effectively reduces engineering risk, reduces the dam accident.Can say, Dam Deformation Monitoring is an important content in engineering survey field.The data of what is more important utilization monitoring are made prediction to relating to safe development of deformation trend, take measures in advance, prevent the generation of security incident.
At present, prediction dam vertical deformation metering method mainly concentrate on theoretical method based on Settlement Mechanism, based on the random statistical method of statistical theory with based on 3 aspects of method of artificial intelligence.3 kinds of method research angles are different, and prediction principle also has basic difference, yet all played preferably effect in Forecast Settlement.But sum up, exist following deficiency:
(1) theoretical method is to be based upon on the detailed engineering geological investigation basis, integrated structure internal characteristics and associated external influence factor are carried out computational analysis, require clear and definite, all kinds of parameter of Settlement Mechanism accurate, but owing to affecting the various complexity of factor of sedimentation, Settlement Mechanism difficulty is understood fully, all kinds of parameters are because of test condition, sampling method restriction also more difficult Obtaining Accurate, so that theoretical method can only be described from single factor based on ideal hypothesis or experience conclusion settlement prediction, thereby cause Calculation results and actual conditions existence than big difference;
(2) the random statistical method is based on measured data, have the advantages such as quick, that calculating is simple, but in modeling process, can only set up the funtcional relationship between settling amount and the single or a plurality of factor, can not take into full account all kinds of influence factors relation each other, be difficult to adapt to the settlement prediction of complex condition;
(3) for artificial intelligence approach, although comparatively be fit to the prediction of sedimentation nonlinear system, calculation of complex, computing time is long, and result of calculation is unstable.
For the above deficiency that these methods exist, stability and the advantage such as combination technique of the present invention by research prediction vertical deformation metering method proposes to adopt certain workflow to come accurately predicting dam dam body vertical deformation amount.Essence of the present invention is that above-mentioned three kinds of methods flow process is according to the rules carried out organically combination, has realized the mutual supplement with each other's advantages of above the whole bag of tricks, can greatly improve the precision of prediction of dam vertical deformation amount.By the actual monitoring data at scene, analyze deformation characteristics and the Changing Pattern thereof of monitoring target, utilize the data of monitoring to make accurately predicting to relating to safe development of deformation trend, take measures in advance, significant to preventing the dam safety accident.
Summary of the invention
Goal of the invention: for the problem and shortage of above-mentioned existing existence, the purpose of this invention is to provide a kind of method of accurately predicting dam dam body vertical deformation amount, the precision of prediction of the method prediction dam deflection is higher, use is more convenient.
Technical scheme: for achieving the above object, the present invention by the following technical solutions: a kind of method of accurately predicting dam dam body vertical deformation amount may further comprise the steps:
The first step, Monitoring Data is divided into two parts in chronological order: a part is time learning data M the preceding, and the number of M must be more than or equal to 10; Another part is the check data J of all the other times;
Second step, utilize learning data M to set up the mathematical model of i kind Forecasting Methodology, i=1 wherein, 2 ... m; I 〉=2;
The degree of stability of the 3rd step, the various Forecasting Methodologies of calculating:
A, dam time series monitor value are y
t(t=1,2, Λ n) is according to the different individual event forecast model of m kind that second step is set up, y
It=(i=1,2, be i kind Single model in t learning value or predicted value constantly Lm), by formula (1) calculates i kind Single model at t period error e
It:
e
it=y
t-y
it (1)
By formula (2) calculate the precision A of i kind Single model in the t phase
It:
Wherein, 0≤α<β≤1, the α acquiescence is got 0 value, and the β acquiescence is got 1 value;
Learn if b i kind Single model at first carries out the N phase, then carry out the T phase and predict, by formula (3)
Calculate the average study precision ε of i kind model
i, the consensus forecast precision η of i kind model
i:
C, then calculates i kind model stability degree S by formula (4)
i:
σ is infinitely small arbitrarily;
The 4th goes on foot, determines combined method model weight coefficient:
Make w
iBe i kind model shared weight in m kind Single model, then the built-up pattern weight coefficient is defined as by formula (5):
The 5th step, according to formula (6), set up the dam vertical deformation amount forecast model based on degree of stability:
In the formula, y
tBe t dam body vertical deformation amount predicted value constantly, w
iBe the weight of i kind model in the built-up pattern, y
ItIt is i kind model t dam body vertical deformation amount predicted value constantly.
As preferably, the mathematical model of the described Forecasting Methodology of second step comprises stepwise regression method forecast model, grey GM(1,1) method forecast model, BP neural net method forecast model.
Beneficial effect: compared with prior art, the present invention has the following advantages: 1, the precision of prediction of dam vertical deformation amount is high, so that the prediction period of Dam Deformation Monitoring enlarges.Through a large amount of case history Analysis of application result, the present invention is at forecast period, and than stepwise regression method, gray method and BP neural net method, the precision that predicts the outcome improves respectively 73.1%, 90.6% and 26.4%.After precision of prediction improved, it is real true that the predicting the outcome of dam vertical deformation amount approaches more, more obvious to dam safety operation and the guiding value that scents a hidden danger; 2, economic benefit is obvious.The present invention can keep the continuity of model accuracy, makes accurately predicting to relating to safe development of deformation trend, takes measures in advance, and is significant to preventing the dam safety accident.Have obvious society and economic worth.
Embodiment
Below in conjunction with specific embodiment, further illustrate the present invention, should understand these embodiment only is used for explanation the present invention and is not used in and limits the scope of the invention, after having read the present invention, those skilled in the art all fall within the application's claims limited range to the modification of the various equivalent form of values of the present invention.
When describing specific implementation process, in conjunction with certain concrete engineering example, the inventive method is elaborated.
(1) case history background introduction
80 phases of monitoring materials that certain concrete dam monolith dam body is nearly 1 year, adopt multipoint displacement meter to monitor this dam body vertical deformation amount, monitor simultaneously (time, temperature, compressive stress, pore water pressure, pore water head, upper pond level etc.) environment parameter corresponding to this monolith.The measured data of 80 phases is as shown in table 1:
Certain Dam Deformation Monitoring measured data of table 1
(2) Data classification
According to as learning data M, rear 30 issues are according to as check data J with front 50 issues of 80 phase Monitoring Data.
(3) individual event Forecasting Methodology
This project example is selected respectively stepwise regression method, grey GM(1,1) method, BP neural net method etc. be as the individual event Forecasting Methodology.
1) stepwise regression method forecast model
This Dam Deformation Monitoring measured data is carried out stepwise regression analysis, final Environment variable: 1. time (D)/sky, 2. temperature (T)/° C, 3. pore water head (H)/m.
Set up the deflection forecast model:
y=a
0+a
1·D+a
2·T+a
3·H (7)
In the formula, the y perpendicular displacement; a
iUndetermined parameter (totally 4) for regression model; D is the cumulative time; T is temperature; H is pore water head absolute altitude.
Specific embodiment with M one by one substitution of Monitoring Data (7) formula, can obtain M error equation, and the error equation general formula is:
v
i=a
0+a
1·D
i+a
2·T
i+a
3·H
i-y
i
Being write as matrix form is:
By " least square method " principle, can obtain the estimated value of 4 undetermined parameter X in the formula (7):
Specific embodiment, one by one substitution of Monitoring Data (7) formula with 50 learning datas (front 50 points in the table 1) can obtain 50 error equations, according to " least square method " principle, obtained the estimated value of 4 undetermined parameters in the formula (7) by formula (9), the results are shown in Table 2.
Table 2 undetermined parameter a
iThe result of calculation table
a 0=-56.281 | a 1=0.055 | a 2=0.390 | a 3=0.528 |
[0059]Successive Regression model data result is as shown in table 3:
Table 3a Successive Regression model data result (unit: mm)
Successive Regression model data result precision is shown in table 3b:
Table 3b Successive Regression model data result accuracy table
2) grey GM(1,1) the method forecast model
Learning data in the deformation measurement data is set up grey GM(1,1) the model processing.
Specific embodiment is with 50 learning datas (front 50 points in the table 1), through grey GM(1,1) to process, Time Created, response function was:
y
t+1=263.016881e
0.026231t-261.966881
Grey GM(1,1) the model data result is as shown in table 4:
Table 4 grey GM(1,1) model data result (unit: mm)
Grey GM(1,1) precision of model data result is as shown in table 5:
Table 5 grey GM(1,1) model data result accuracy table
3) BP neural net method forecast model
Learning data in the deformation measurement data is set up the BP neural network model.Specific embodiment carries out the training of BP neural network model with 50 learning datas (front 50 points in the table 1), and parameter configuration is as follows:
Model structure: 6 * 18 * 1.That is: the input layer number of parameters of network is 6, and 6 input parameters are respectively: time/sky, temperature/℃, compressive stress/Mpa, pore water pressure/Kpa, pore water head/m and upper pond level/m; The hidden layer number of parameters of network is 18; The output layer number of parameters of network is 1, is dam body vertical deformation amount y
It
BP neural network model data processed result is as shown in table 6:
Table 6BP neural network model data processed result (unit: mm)
The precision of BP neural network model data processed result is as shown in table 7:
Table 7BP neural network model data processed result accuracy table
4) based on degree of stability combined weights parameter identification
According to determining method based on the degree of stability weight coefficient, select Successive Regression model, grey GM(1,1) model and BP neural network model be as single model, sets up built-up pattern.Set up and data analysis calculating through model, choice accuracy factor-alpha and β are default value, and namely α gets 0 value, and β gets 1 value.The weight coefficient that gets built-up pattern according to formula (1), formula (2), formula (3), formula (4) and formula (5) is:
w
1=0.275703,w
2=0.048617,w
3=0.675680
In this built-up pattern, w
1Be the weight coefficient of Successive Regression model, w
2Be grey GM(1,1) weight coefficient of model, w
3Weight coefficient for the BP neural network model.
Obviously, w
1+ w
2+ w
3=1,0≤w
i≤ 1, i=1,2,3.
5) foundation is based on the dam vertical deformation amount forecast model of degree of stability
The dam vertical deformation amount forecast model of setting up according to formula (6) is:
y
t=0.275703y
1t+0.048617y
2t+0.675680y
3t
Built-up pattern data processed result based on degree of stability power is as shown in table 8:
Table 8 built-up pattern data processed result (unit: mm)
The precision of built-up pattern data processed result of determining method based on the degree of stability weight coefficient is as shown in table 9:
Table 9 is determined the built-up pattern data processed result accuracy table of method based on the degree of stability weight coefficient
This example has 50 learning samples, and 30 checks (prediction) sample utilizes forecast sample can estimate the prediction effect of distinct methods.Valuation with standard deviation
Estimate its precision:
In the formula,
Be predicting the outcome of t phase distinct methods, y
tBe the measured value of t phase, n is the number of check point.Definition by standard deviation is known, the valuation of the standard deviation of check point
Less, precision is higher, shows that prediction effect is better.Assay sees Table 10.Compare with the BP neural network that precision is best, the precision of the inventive method prediction dam body vertical deformation amount can improve 26.4%.
Table 10 distinct methods testing accuracy comparison sheet
。
Claims (2)
1. the method for an accurately predicting dam dam body vertical deformation amount may further comprise the steps:
The first step, Monitoring Data is divided into two parts in chronological order: a part is time learning data M the preceding, and the number of M must be more than or equal to 10; Another part is the check data J of all the other times;
Second step, utilize learning data M to set up the mathematical model of i kind Forecasting Methodology, i=1 wherein, 2 ... m; I 〉=2;
The degree of stability of the 3rd step, the various Forecasting Methodologies of calculating:
A, dam time series monitor value are y
t(t=1,2, Λ n) is according to the different individual event forecast model of m kind that second step is set up, y
It=(i=1,2, be i kind Single model in t learning value or predicted value constantly Lm), by formula (1) calculates i kind Single model at t period error e
It:
e
it=y
t-y
it (1)
By formula (2) calculate the precision A of i kind Single model in the t phase
It:
Wherein, 0≤α<β≤1, the α acquiescence is got 0 value, and the β acquiescence is got 1 value;
Learn if b i kind Single model at first carries out the N phase, then carry out the T phase and predict, by formula (3) calculate the average study precision ε of i kind model
i, the consensus forecast precision η of i kind model
i:
C, then calculates i kind model stability degree S by formula (4)
i:
σ is infinitely small arbitrarily;
The 4th goes on foot, determines combined method model weight coefficient:
Make w
iBe i kind model shared weight in m kind Single model, then the built-up pattern weight coefficient is defined as by formula (5):
The 5th step, according to formula (6), set up the dam vertical deformation amount forecast model based on degree of stability:
In the formula, y
tBe t dam body vertical deformation amount predicted value constantly, w
iBe the weight of i kind model in the built-up pattern, y
ItIt is i kind model t dam body vertical deformation amount predicted value constantly.
2. the method for described accurately predicting dam dam body vertical deformation amount according to claim 1, it is characterized in that: the mathematical model of the described Forecasting Methodology of second step comprises stepwise regression method forecast model, grey GM(1,1) method forecast model, BP neural net method forecast model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310189139.0A CN103353295B (en) | 2013-05-20 | 2013-05-20 | A kind of method of accurately predicting dam dam body vertical deformation amount |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310189139.0A CN103353295B (en) | 2013-05-20 | 2013-05-20 | A kind of method of accurately predicting dam dam body vertical deformation amount |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103353295A true CN103353295A (en) | 2013-10-16 |
CN103353295B CN103353295B (en) | 2015-11-18 |
Family
ID=49309689
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310189139.0A Expired - Fee Related CN103353295B (en) | 2013-05-20 | 2013-05-20 | A kind of method of accurately predicting dam dam body vertical deformation amount |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103353295B (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942430A (en) * | 2014-04-21 | 2014-07-23 | 南京市测绘勘察研究院有限公司 | Building settlement prediction method based on combined model |
CN105868556A (en) * | 2016-03-29 | 2016-08-17 | 武汉福天通科技有限公司 | Bridge safety assessment method based on Beidou bridge safety monitoring system |
CN105956662A (en) * | 2016-04-20 | 2016-09-21 | 东南大学 | Underground structure deformation prediction method based on BP-regression analysis prediction model |
CN106918365A (en) * | 2017-03-28 | 2017-07-04 | 深圳智达机械技术有限公司 | A kind of reliability Monitoring System for Dam Safety high |
CN110287634A (en) * | 2019-07-03 | 2019-09-27 | 中国水利水电科学研究院 | A kind of dam abutment deformation analogy method applied based on body force |
CN110374047A (en) * | 2019-05-28 | 2019-10-25 | 中国水利水电科学研究院 | Arch dam runtime real-time security monitoring Threshold based on deformation |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
SU1191513A1 (en) * | 1984-02-17 | 1985-11-15 | Горьковский Ордена Трудового Красного Знамени Инженерно-Строительный Институт Им.В.П.Чкалова | Dam made of soil materials |
CN101699477A (en) * | 2009-10-21 | 2010-04-28 | 东南大学 | Neural network method for accurately predicting dam deformation |
CN103292762A (en) * | 2013-05-23 | 2013-09-11 | 青岛理工大学 | Displacement monitoring method for judging dam stability |
-
2013
- 2013-05-20 CN CN201310189139.0A patent/CN103353295B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
SU1191513A1 (en) * | 1984-02-17 | 1985-11-15 | Горьковский Ордена Трудового Красного Знамени Инженерно-Строительный Институт Им.В.П.Чкалова | Dam made of soil materials |
CN101699477A (en) * | 2009-10-21 | 2010-04-28 | 东南大学 | Neural network method for accurately predicting dam deformation |
CN103292762A (en) * | 2013-05-23 | 2013-09-11 | 青岛理工大学 | Displacement monitoring method for judging dam stability |
Non-Patent Citations (2)
Title |
---|
张黎明等: "基于卡尔曼滤波法的碾压混凝土拱坝垂直位移变形分析", 《水电自动化与大坝监测》, vol. 35, no. 6, 20 December 2011 (2011-12-20) * |
徐培亮: "应用时间序列方法作大坝变形预报", 《武汉测绘科技大学学报》, vol. 13, no. 3, 30 September 1988 (1988-09-30) * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103942430A (en) * | 2014-04-21 | 2014-07-23 | 南京市测绘勘察研究院有限公司 | Building settlement prediction method based on combined model |
CN103942430B (en) * | 2014-04-21 | 2017-08-11 | 南京市测绘勘察研究院有限公司 | A kind of building settlement Forecasting Methodology based on built-up pattern |
CN105868556A (en) * | 2016-03-29 | 2016-08-17 | 武汉福天通科技有限公司 | Bridge safety assessment method based on Beidou bridge safety monitoring system |
CN105956662A (en) * | 2016-04-20 | 2016-09-21 | 东南大学 | Underground structure deformation prediction method based on BP-regression analysis prediction model |
CN106918365A (en) * | 2017-03-28 | 2017-07-04 | 深圳智达机械技术有限公司 | A kind of reliability Monitoring System for Dam Safety high |
CN110374047A (en) * | 2019-05-28 | 2019-10-25 | 中国水利水电科学研究院 | Arch dam runtime real-time security monitoring Threshold based on deformation |
CN110287634A (en) * | 2019-07-03 | 2019-09-27 | 中国水利水电科学研究院 | A kind of dam abutment deformation analogy method applied based on body force |
Also Published As
Publication number | Publication date |
---|---|
CN103353295B (en) | 2015-11-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101699477B (en) | Neural network method for accurately predicting dam deformation | |
Huang et al. | Monthly streamflow prediction using modified EMD-based support vector machine | |
CN105139585B (en) | A kind of soil-slope dangerous situation intelligent early-warning forecasting procedure | |
CN103353295B (en) | A kind of method of accurately predicting dam dam body vertical deformation amount | |
CN104318077B (en) | Climate change and mankind's activity change quantitative analysis method to river flow | |
CN109492823B (en) | Method for predicting icing thickness of power transmission line | |
CN111191191B (en) | Construction method of combined model for accurately predicting deformation effect of concrete dam | |
CN104794361A (en) | Comprehensive evaluation method for water flooding oil reservoir development effect | |
Castellani et al. | Wind energy forecast in complex sites with a hybrid neural network and CFD based method | |
CN103646181A (en) | Method for determining wriggle slide type artificial side slope stability coefficient and early warning criteria | |
CN111178758A (en) | Concrete dam monitoring data intelligent management and real-time evaluation system based on BIM | |
CN107292022A (en) | A kind of bridge structure probability baseline finite element model construction method responded based on time varying temperature | |
CN103439070A (en) | Separation method for long-term deflection effect of bridge | |
CN104506162A (en) | Fault prognosis method for high-order particle filter on basis of LS-SVR (least squares support vector regression) modeling | |
Wang et al. | Forecast of water level and ice jam thickness using the back propagation neural network and support vector machine methods | |
CN103530818A (en) | Water supply pipe network modeling method based on BRB (belief-rule-base) system | |
Qiu et al. | Seepage monitoring models study of earth-rock dams influenced by rainstorms | |
CN105824987A (en) | Wind field characteristic statistical distributing model building method based on genetic algorithm | |
CN103134433A (en) | Method of identifying slip factors causing slope instability by utilizing displacement monitoring | |
Yang et al. | Comparative Study on Deformation Prediction Models of Wuqiangxi Concrete Gravity Dam Based onMonitoring Data. | |
CN103279030B (en) | Dynamic soft measuring modeling method and device based on Bayesian frame | |
CN103886177A (en) | Lightning fire daily occurrence probability predicting method based on space grids | |
CN103226737A (en) | Chemical abnormal condition trend prediction method based on kalman filtering and grey prediction | |
CN104102853A (en) | Slope displacement fractal forecasting method improved by grey theory | |
Tang et al. | Displacement prediction of rainfall induced landslide based on AdaBoost BP neural network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20151118 Termination date: 20180520 |