Application of an artificial immune algorithm on a statistical model of dam displacement

Xi, Guangyong; Yue, Jianping; Zhou, Baoxing; Pu, Tao

doi:10.1016/j.camwa.2011.09.057

Cited by 41 publications

(18 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Loh et al applied two different approaches to the continuous monitoring data of static deformation of the Fei‐Tsui Arch Dam (Taiwan) to extract the tendencies of the deformation and determine the threshold level for early warning on dam static deformation. Xi et al established a new statistical model by using the artificial immune algorithm to solve the data analysis problem of dam crest displacement and predict the future behaviors of the dam. Xu et al proposed a statistically inspired modification of the PLS regression algorithm with the predictor variables selected by the genetic algorithm with partial least squares to cope with the multicollinearity problem of dam regression models.…”

Section: Introductionmentioning

confidence: 99%

Concrete dam deformation prediction model for health monitoring based on extreme learning machine

Kang

Liu

et al. 2017

Struct. Control Health Monit.

195

View full text Add to dashboard Cite

Summary Structural health monitoring via quantities that can reflect behaviors of concrete dams, like horizontal and vertical displacements, rotations, stresses and strains, seepage, and so forth, is an important method to evaluate operational states of concrete dams correctly and predict the future structural behaviors accurately. Traditionally, statistical model is widely applied in practical engineering for structural health monitoring. In this paper, an extreme learning machine (ELM)‐based health monitoring model is proposed for displacement prediction of gravity dams. ELM is one type of https://en.wikipedia.org/wiki/Feedforward_neural_networks with a single layer of hidden nodes, where the weights connecting inputs to hidden nodes are randomly assigned. The model can produce good generalization performance and learns faster than networks trained using the back propagation algorithm. The advantages such as easy operating, high prediction accuracy, and fast training speed of the ELM health monitoring model are verified by monitoring data of a real concrete dam. Results are also compared with that of the back propagation neural networks, multiple linear regression, and stepwise regression models for dam health monitoring.

show abstract

Section: Introductionmentioning

confidence: 99%

Concrete dam deformation prediction model for health monitoring based on extreme learning machine

Kang

Liu

et al. 2017

Struct. Control Health Monit.

195

View full text Add to dashboard Cite

show abstract

“…Dam deformation is driven by internal and external environment factors, including water load, temperature load, structural damage, seepage coupling, and joint fissure. With deepening recognition of the complexity of dam deformation, artificial neural network, time series analysis, extreme learning machine, artificial immune algorithm, support vector machine, system optimization model, and many other algorithms have been used in dam deformation statistical analysis. In short, the above‐mentioned methods to establish the statistical model can be summarized as two methods, the HST (hydrostatic, seasonal, time) model and the HTT (hydrostatic, thermal, time) model .…”

Section: Introductionmentioning

confidence: 99%

Modified hybrid forecast model considering chaotic residual errors for dam deformation

Wei

Yuan

et al. 2018

Struct Control Health Monit

View full text Add to dashboard Cite

Summary Dam deformation is double effected by the internal and external environments, showing high nonlinear characteristics. Information mining of dam prototype observation data and effective components contained in the residual sequence is limited in conventional dam deformation forecast models. In order to fully explore the complex nonlinear relationship between dam deformation and environmental factors, modified hybrid forecast model considering chaotic residual errors is proposed on the combination of shuffled frog leapfrog algorithm and chaos theory in this study. Hybrid models are established on the basis of dam deformation prototype observation data, and modified hybrid forecast model is established by determining the optimal weight of each hybrid model with shuffled frog leapfrog algorithm. Considering the chaotic characteristics contained in the residual sequence of modified hybrid forecast model, residual sequence is analyzed and forecasted by chaos theory. By superimposing the residual forecast value with the forecast value of modified hybrid forecast model, modified hybrid forecast model considering chaotic residual errors is established. Example shows that, compared with conventional models, the proposed model is better in fitting precision and convergence speed, and forecast capability is significantly improved by considering the effective components contained in the residual sequence, which presents a new method of the deformation forecast for other hydraulic structures.

show abstract

“…Another important issue is data reduction from hundreds of monitored instruments and the identification of critical parameters of dam responses. Previous studies have utilized principal component analysis,() artificial neural networks,() blind source separation,() artificial immune algorithm,() independent component regression,() cointegration theory,() and time‐varying Bayesian approach() to model and predict the dam responses. Time‐frequency analysis (short‐time Fourier transform) was employed by Mata et al() to identify the effect of daily variations of air temperature on the structural response of a concrete dam.…”

Section: Introductionmentioning

confidence: 99%

Initial service life data towards structural health monitoring of a concrete arch dam

Prakash

Sadhu

Narasimhan

et al. 2017

Struct Control Health Monit

View full text Add to dashboard Cite

SummaryThis paper presents a statistical framework to monitor the performance of an operational concrete arch dam using sensory data acquired during its initial service life.One of the major challenges in dealing with a newly constructed dam is to predict its long-term behaviour by forecasting appropriate thresholds using limited data exhibiting nonstationarity. In this paper, a hybrid model is implemented to predict dam responses using environmental-hydrostatic, seasonal, and temperature-as well as age-related variables. The data from multiple sensors are first analyzed using principal component analysis to incorporate overall dam behaviour into a prediction model. The proposed prediction framework is then employed to estimate the residuals and control limits required to calculate thresholds under nonstationary operating conditions during its initial service life. The dam performance is then monitored using statistical control charts and anomalies are detected by comparing the test statistics, square prediction error, and Hotelling T-squared, calculated from the residuals with the preset control limits. The issue of limited data is addressed by updating the model parameters and thresholds periodically, which is aimed at minimizing the false alarm rate. The proposed method is demonstrated using a 130-m-high double-arch concrete dam located in Bulgaria.

show abstract

Application of an artificial immune algorithm on a statistical model of dam displacement

Cited by 41 publications

References 12 publications

Concrete dam deformation prediction model for health monitoring based on extreme learning machine

Concrete dam deformation prediction model for health monitoring based on extreme learning machine

Modified hybrid forecast model considering chaotic residual errors for dam deformation

Initial service life data towards structural health monitoring of a concrete arch dam

Contact Info

Product

Resources

About