Performance improvement method of support vector machine-based model monitoring dam safety

Abstract

Summary Under the comprehensive influence of material and loads, dam structural behavior presents the time‐varying nonlinear characteristics. To forecast the dam structural behavior (displacement, stress, seepage, etc.), the models monitoring dam safety are often built according to the prototype observations on dam safety. However, the modeling process is usually fulfilled with the offline and static pattern. As time goes on, the fitting and forecasting ability of built static model will decline gradually. The article is focused on the support vector machine (SVM)‐based model monitoring dam safety. The methods are studied to advance the adaptability of SVM model and reduce the modeling time. By implementing the impact analysis for SVM parameters and input vector, the optimization method of SVM parameters and input vector is presented to enhance the efficiency of building the SVM‐based static model monitoring dam safety. To describe dynamically the time‐varying mapping relationship between dam structural behavior (effect‐quantity) and its cause (influence‐quantity), the way is developed to update in real time above model by making the most use of new observations. The displacement of one actual dam is taken as an example to verify the modeling efficiency and forecasting ability. Copyright © 2015 John Wiley & Sons, Ltd.

Linked Simulation-Optimization Model for Optimum Hydraulic Design of Water Retaining Structures Constructed on Permeable Soils

“…Recently, SVM has been utilized to model complex civil engineering problems to provide an extensive understanding of the variables involved in the model [9][10][11][12][13][14][15][16].In the majority of the conducted researches, SVM is utilized to predict specific responses depending on the input variables or to simulate performance of a particular engineering system with special conditions. For example, in hydraulic structure design, SVM has been used to predict the forecasting of the tangential shift of a concrete dam [14], and to predict future dam responses with environmental variables [17].…”

Multisource information fusion-based approach diagnosing structural behavior of dam engineering

“…First, the data series of single monitoring point is taken to analyze its characteristic values and build its mathematical model with Kalman filter method, support vector machine method, and artificial neural network method. Then the information fusion for monitoring points with the same type is implemented to construct the space–time model describing quantitatively dam structural behavior …”

Modeling method for predicting seepage of RCC dams considering time-varying and lag effect

“…Thus, k v can be approximately regarded as the permeability coefficient of the RCC dam when the layer does not exist. According to the literature, the relationship curve between the measuring head and the reservoir water level at the measuring point under different reservoir water levels is obtained by the finite element analysis method of seepage. The following polynomial is obtained by a further fitting solution: where a i is the fitting coefficient of water pressure, q 1 is the number of factors, and H is the upstream water depth.…”

“…We carried out a basic research on the mechanical characteristics of RCC dams and found that their interlayer seepage characteristics are different from that of the RCC noumenon, and the effect zone is an important factor in the percolation state analysis of RCC dams. The permeability of RCC dams is influenced by factors, such as reservoir level, rainfall, temperature, and aging . According to the seepage monitoring data analysis, the influence of the reservoir water level and rainfall on the seepage of dams has a lag effect .…”