Support vector regression based metamodeling for structural reliability analysis

Roy, Atin; Manna, Ramkrishna; Chakraborty, Subrata

doi:10.1016/j.probengmech.2018.11.001

Cited by 94 publications

(31 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Big hydrological data contain the hydrological evolution pattern; recently, the introduction of artificial intelligence methods to the hydrological data prediction has gained much attention, and many artificial intelligence relevance methods have been applied to hydrological prediction, such as neural networks, [5][6][7][8] fuzzy theory, [9][10][11] and genetic algorithms, [12][13][14] which have greatly promoted the development of rainfall prediction. Support vector machine (SVM) [15][16][17] holds the advantage of highly generalized and is widely used in machine learning context, while its prediction result is biased due to the incomplete hydrological data. Kalman filter (EnKF) 18,19 can well handle the nonlinear hydrological data, but it requires the rather lengthy historical data.…”

Section: Related Workmentioning

confidence: 99%

Hybrid model of generative adversarial network and Takagi‐Sugeno for multidimensional incomplete hydrological big data prediction

Liu

Song

2020

Concurrency and Computation

View full text Add to dashboard Cite

Summary The processing of rainfall‐runoff transmission in the catchment area is a very complex phenomenon, such as temporal and spatial changes of the catchment characteristics and uncertainties of rainfall patterns. To handle this challenge, data‐driven hydrologic models emerge rapidly for the rainfall runoff prediction. However, the incomplete hydrological data constrain the development of digital hydrologic model. This article proposes a rainfall‐runoff prediction method of coupling generative adversarial network (GAN) and Takagi‐Sugeno (T‐S) fuzzy model, in which the GAN is used to generate the data for completing the incomplete hydrological data, and the T‐S fuzzy model is utilized for forecasting the rainfall runoff. The presented method is examined by real data in Huaihe basin just considering the precipitation and streamflow. Experiments show that the model combined GAN and T‐S can achieve satisfactory prediction results.

show abstract

Section: Related Workmentioning

confidence: 99%

Hybrid model of generative adversarial network and Takagi‐Sugeno for multidimensional incomplete hydrological big data prediction

Liu

Song

2020

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Namun, SVM dikembangkan sehingga dapat digunakan untuk memprediksi kuantitas dari suatu amatan atau dikenal dengan pemodelan regresi (Vapnik, 2000). Support Vector Rergression dengan kernel Radial Basis Function (SVR-RBF) merupakan metode yang baik digunakan dalam pemodelan regresi nonlinier (Ma et al, 2018;Pour et al, 2016;Roy et al, 2019). Karimi et al (2008) menunjukkan bahwa SVR-RBF dapat mengatasi masalah prediktor berkorelasi tinggi yang ada dalam suatu gugus data riil atau empiris.…”

Section: Pendahuluanunclassified

Kajian Simulasi Perbandingan Metode Regresi Kuadrat Terkecil Parsial, Support Vector Machine, Dan Random Forest

Fauzi

Soleh

Djuraidah

2020

IJSA

View full text Add to dashboard Cite

Highly correlated predictors and nonlinear relationships between response and predictors potentially affected the performance of predictive modeling, especially when using the ordinary least square (OLS) method. The simple technique to solve this problem is by using another method such as Partial Least Square Regression (PLSR), Support Vector Regression with kernel Radial Basis Function (SVR-RBF), and Random Forest Regression (RFR). The purpose of this study is to compare OLS, PLSR, SVR-RBF, and RFR using simulation data. The methods were evaluated by the root mean square error prediction (RMSEP). The result showed that in the linear model, SVR-RBF and RFR have large RMSEP; OLS and PLSR are better than SVR-RBF and RFR, and PLSR provides much more stable prediction than OLS in case of highly correlated predictors and small sample size. In nonlinear data, RFR produced the smallest RMSEP when data contains high correlated predictors.

show abstract

“…Pan and Dias (2017) developed an efficient reliability method using an adaptive SVM, for which informative training samples are sequentially selected through a learning function. Roy et al (2019) proposed a metamodeling method based on support vector regression for structural reliability analysis. Zhu and Du (2016) proposed a new Kriging-based active learning reliability method that considers the dependencies between Kriging predictions.…”

Section: Introductionmentioning

confidence: 99%

Efficient sampling-based inverse reliability analysis combining Monte Carlo simulation (MCS) and feedforward neural network (FNN)

Lee

2021

Struct Multidisc Optim

View full text Add to dashboard Cite

Inverse reliability analysis evaluates a percentile value of a performance function when the target reliability is given. In cases of high dimensional or highly nonlinear performance functions, sampling-based methods such as Monte Carlo simulation (MCS), Latin hypercube sampling, and importance sampling are considered to be better candidates for reliability analysis. The sampling-based methods are very accurate but require a large number of samples, which can be very time-consuming. Therefore, this paper proposes an efficient and/or accurate sampling-based reliability analysis method without using a surrogate model. The proposed method helps to improve the accuracy of reliability analysis with the same number of samples or to ensure the same accuracy of reliability analysis with fewer samples. This study starts with an idea of training relationship between limited samples constituting realization of the performance distribution-usually between 10 and 100-and its corresponding true percentile value where the performance distribution is defined as a one-dimensional distribution resulted from a performance function and its random variables. To this end, feedforward neural network (FNN), which is one of promising artificial neural network (ANN) models that approximate high dimensional models using layered structures, is introduced in this study, and limited samples constituting realizations of various performance distributions and their corresponding true percentile values are used as input and target data, respectively. Various beta distributions are used to create the training data sets. A FNN training method using kernel density estimation and equidistant points to represent the kernel distribution data is also proposed to remove dimensionality of the training inputs. Comparative study shows that the proposed method training FNN with samples constituting realization of the performance distribution (Method 2) is more accurate than a method that directly estimates the percentile value from the kernel distribution fitting the samples constituting realization generated through MCS (Method 1). In addition, compared to Method 2, another proposed method that trains FNN with the kernel density estimation and equidistant points (Method 3) is more accurate in reliability analysis and more computationally efficient in FNN training. Method 3 is also applicable to high reliability problems, and it is more accurate than Kriging-based method for high dimensional problems.

show abstract

Support vector regression based metamodeling for structural reliability analysis

Cited by 94 publications

References 40 publications

Hybrid model of generative adversarial network and Takagi‐Sugeno for multidimensional incomplete hydrological big data prediction

Hybrid model of generative adversarial network and Takagi‐Sugeno for multidimensional incomplete hydrological big data prediction

Kajian Simulasi Perbandingan Metode Regresi Kuadrat Terkecil Parsial, Support Vector Machine, Dan Random Forest

Efficient sampling-based inverse reliability analysis combining Monte Carlo simulation (MCS) and feedforward neural network (FNN)

Contact Info

Product

Resources

About