AK-ILS: An Active learning method based on Kriging for the Inspection of Large Surfaces

Dumas, Antoine; Echard, Benjamin; Gayton, Nicolas; Rochat, Olivier; Dantan, Jean‐Yves; Veen, Sjoerd van der

doi:10.1016/j.precisioneng.2012.07.007

Cited by 31 publications

(15 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since there is only a single data source, the multi-dataset regression becomes a single GP modelling. It has been noted in published literature [ 18 , 32 ] that GP based adaptive sampling has superior performance than other sampling methods in sampling complex surfaces. Hence, the results are also used as benchmarking to examine the efficiency of the proposed GP-BIS.…”

Section: Experimental Studymentioning

confidence: 99%

“…Recently, Gaussian process (GP) has been used to address the multi-sensor data fusion problem [ 29 ]. GP is kind of a non-parametric Bayesian inference model which has been proved a powerful mathematical tool to model various complex surfaces in terms of spatial covariance functions [ 30 , 31 , 32 ]. The statistic nature of GP makes it capable of incorporating the measurement errors into the modelling process and assigning credibility to the constructed model.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Gaussian Process Based Bayesian Inference System for Intelligent Surface Measurement

Ren

Cheung

Xiao

2018

Sensors

View full text Add to dashboard Cite

This paper presents a Gaussian process based Bayesian inference system for the realization of intelligent surface measurement on multi-sensor instruments. The system considers the surface measurement as a time series data collection process, and the Gaussian process is used as mathematical foundation to establish an inferring plausible model to aid the measurement process via multi-feature classification and multi-dataset regression. Multi-feature classification extracts and classifies the geometric features of the measured surfaces at different scales to design an appropriate composite covariance kernel and corresponding initial sampling strategy. Multi-dataset regression takes the designed covariance kernel as input to fuse the multi-sensor measured datasets with Gaussian process model, which is further used to adaptively refine the initial sampling strategy by taking the credibility of the fused model as the critical sampling criteria. Hence, intelligent sampling can be realized with consecutive learning process with full Bayesian treatment. The statistical nature of the Gaussian process model combined with various powerful covariance kernel functions offer the system great flexibility for different kinds of complex surfaces.

show abstract

Section: Experimental Studymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Gaussian Process Based Bayesian Inference System for Intelligent Surface Measurement

Ren

Cheung

Xiao

2018

Sensors

View full text Add to dashboard Cite

show abstract

“…Their dimensions cannot be described by a single number. In order to improve dimensional accuracy of free-form surfaces parts, the existing researches have mainly focused on the free-form surfaces machining methods [21][22][23] and measurement methods [24][25][26][27][28]. In fact, dimensional deviation estimation belongs to the mean value estimation problem.…”

Section: Mathematical Problems In Engineeringmentioning

confidence: 99%

Dimensional Deviation Estimation for Parts with Free-Form Surfaces

Zhao

Wang

Zhou

et al. 2018

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Dimensional deviation is a prerequisite for improving the manufacturing process of parts with free-form surfaces, for example, the reverse adjustment of the die cavity of turbine blade. Influenced by random noise of the manufacturing process, dimensional variation is inevitable for batch parts. Therefore, it may cause unacceptable error to estimate batch parts dimensional deviation by a single sample. Meanwhile, the optimum sample size for estimating dimensional deviation is difficult to determine. To overcome this problem, a practical method for estimating of dimensional deviation of parts with free-form surface is proposed. Firstly, displacements of the discrete points on part surface are employed to represent dimensional error of the part. Estimating dimensional deviation of parts is actually to estimate the simultaneous confidence intervals of the discrete point displacements. Secondly, Bonferroni simultaneous confidence intervals are adopted to estimate the confidence intervals of the part dimensional deviation. Moreover, the accuracy of the estimation will be continuously improved by increasing samples. Consequently, a practical dimensional deviation estimation method is presented. Finally, a compressor blade is adopted to illustrate the proposed method. The percentage of estimation error of the blade dimensional deviation that is less than 0.05mm, which is the estimation error limit, of all the 100 times sampling experiments is 99%, exceeding the given confidence level of 95%, while the percentage of the existing method is below the confidence level, only 87%.

show abstract

“…To reduce the computational cost, an ALK method [43,47,48,[66][67][68] is introduced to approximate the performance in the region of interest rather than throughout the overall space. Based on the Kriging model, then, the failure probability U f can be calculated through MCS method.…”

Section: Fundamental Theorymentioning

confidence: 99%

“…(a) The number of training points included in the initial DOE is supposed to be small. The number of 12 is selected in this study, according to the existing investigations [47,48]. (b) Latin hypercube sampling (LHS) is utilized to produce the samples, which are uniformly distributed in the uncertain space.…”

Section: Summary Of Alk-mghar-mcs For Reliability Analysismentioning

confidence: 99%

A New Formulation on Seismic Risk Assessment for Reinforced Concrete Structures with Both Random and Bounded Uncertainties

Liu

Wang

2018

Discrete Dynamics in Nature and Society

View full text Add to dashboard Cite

A new formulation on seismic risk assessment for structures with both random and uncertain-but-bounded variables is investigated in this paper. Limit thresholds are regarded as random variables. The median of random variables is described through an improved multidimensional parallelepiped (IMP) convex model, in which the uncertain domain of the dependent bounded variables can be explicitly expressed. The corresponding Engineering Demand Parameters are taken to be dependent and follow a multidimensional lognormal distribution. Through matrix transformation, a given performance function is transformed into the regularized one. An effective method based on active learning Kriging model (ALK) is introduced to approximate the performance function in the region of interest rather than in the overall uncertain space. Based on ALK model, the failure probabilities for different limit states are calculated by using Monte Carlo Simulation (MCS). Further, the failure probabilities for different limit states in 50 years can be obtained through coupling the seismic failure probability with the ground motion hazard curve. A six-story reinforced concrete building subjected to ground motions is investigated to the efficiency and accuracy of the proposed method. The interstory drift and the acceleration as two responses of the case study are, respectively, obtained by utilizing Incremental Dynamic Analysis and nonlinear history analysis.

show abstract

AK-ILS: An Active learning method based on Kriging for the Inspection of Large Surfaces

Cited by 31 publications

References 13 publications

Gaussian Process Based Bayesian Inference System for Intelligent Surface Measurement

Gaussian Process Based Bayesian Inference System for Intelligent Surface Measurement

Dimensional Deviation Estimation for Parts with Free-Form Surfaces

A New Formulation on Seismic Risk Assessment for Reinforced Concrete Structures with Both Random and Bounded Uncertainties

Contact Info

Product

Resources

About