The D-optimal minimax criterion is proposed to construct fractional factorial designs. The resulting designs are very efficient, and robust against misspecification of the effects in the linear model. The criterion wa...
详细信息
The D-optimal minimax criterion is proposed to construct fractional factorial designs. The resulting designs are very efficient, and robust against misspecification of the effects in the linear model. The criterion was first proposed by Wilmut & Zhou (2011);their work is limited to two-level factorial designs, however. In this paper we extend this criterion to designs with factors having any levels (including mixed levels) and explore several important properties of this criterion. Theoretical results are obtained for construction of fractional factorial designs in general. This minimax criterion is not only scale invariant, but also invariant under level permutations. Moreover, it can be applied to any run size. This is an advantage over some other existing criteria. The Canadian Journal of Statistics 41: 325340;2013 (c) 2013 Statistical Society of Canada.
Finding the optimal threshold(s) for an image with a multimodal histogram is described in well-known literature as a problem in fitting a sum of Gaussians to the histogram. This fitting problem is shown experimentally...
详细信息
Finding the optimal threshold(s) for an image with a multimodal histogram is described in well-known literature as a problem in fitting a sum of Gaussians to the histogram. This fitting problem is shown experimentally to be a nonlinear minimization with local minima. A new minimization technique, tree annealing, is presented which finds the global minimum. Experimental results for histograms with two and three modes are presented.
M-robust designs are defined and constructed for misspecified linear regression models with possibly autocorrelated errors on a discrete design space. These designs minimize the mean-squared errors if linear regressio...
详细信息
M-robust designs are defined and constructed for misspecified linear regression models with possibly autocorrelated errors on a discrete design space. These designs minimize the mean-squared errors if linear regression models are correct with uncorrelated errors, subject to two robust constraints which control the change of the bias and the change of variance under model departures. Simulated annealing algorithm is applied to construct M-robust designs. Examples are given to show M-robust designs and compare them with minimax robust designs. (C) 2004 Published by Elsevier B.V.
A D-optimal minimax design criterion is proposed to construct two-level fractional factorial designs, which can be used to estimate a linear model with main effects and some specified interactions. D-optimal minimax d...
详细信息
A D-optimal minimax design criterion is proposed to construct two-level fractional factorial designs, which can be used to estimate a linear model with main effects and some specified interactions. D-optimal minimax designs are robust against model misspecification and have small biases if the linear model contains more interaction terms. When the D-optimal minimax criterion is compared with the D-optimal design criterion, we find that the D-optimal design criterion is quite robust against model misspecification. Lower and upper bounds derived for the loss functions of optimal designs can be used to estimate the efficiencies of any design and evaluate the effectiveness of a search algorithm. Four algorithms to search for optimal designs for any run size are discussed and compared through several examples. An annealing algorithm and a sequential algorithm are particularly effective to search for optimal designs. (C) 2010 Elsevier B.V. All rights reserved.
Consider the design problem for the approximately linear model with serially correlated errors. The correlated structure is the qth degree moving average process, MA(q), especially for q = 1, 2. The optimal design is ...
详细信息
Consider the design problem for the approximately linear model with serially correlated errors. The correlated structure is the qth degree moving average process, MA(q), especially for q = 1, 2. The optimal design is derived by using Bayesian approach. The Bayesian designs derived with various priors are compared with the classical designs with respect to some specific correlated structures. The results show that any prior knowledge about the sign of the MA(q) process parameters leads to designs that are considerately more efficient than the classical ones based on homoscedastic assumptions.
As there are great absorption and scattering in water, it is difficult to extract the target region from an underwater image effectively. This paper investigated a Bayesian decision-making framework for segmenting und...
详细信息
As there are great absorption and scattering in water, it is difficult to extract the target region from an underwater image effectively. This paper investigated a Bayesian decision-making framework for segmenting underwater images, with the improved OTSU algorithm combined with the simulated annealing algorithm calculating the optimum threshold. The improved OTSU algorithm took fully into account grey values of pixels and their neighbours to have a better ability of filtering noise. The simulated annealing algorithm was contributed to reduce the amount of calculation and improved the efficiency of calculating the optimum threshold. Blob operators were used to exclude fake target regions based on Bayesian decision-making. The mathematical morphology operators were used to eliminate burrs and disturbances. The result of processing the images grabbed at pool experiments proved the better capability of segmentation with the proposed method.
Minimax robust designs for regression models with possible misspecification in the response and possible autocorrelated errors are investigated on discrete design spaces. The designs minimize the maximum value of the ...
详细信息
Minimax robust designs for regression models with possible misspecification in the response and possible autocorrelated errors are investigated on discrete design spaces. The designs minimize the maximum value of the trace of the mean squared error (MSE) matrix, and the maximum is obtained over a class of departure functions from the regression response and a class of autocorrelated errors. In particular, classes of moving average error processes are studied. The maximum value of the trace of MSE is obtained analytically, and the minimax designs can be computed through an annealing algorithm. Several examples are given to show robust designs for polynomial regression. (c) 2007 Elsevier B.V. All rights reserved.
The optimal contribution selection method and the simulated annealing algorithm were used to maximize the genetic gain and reduce inbreeding in fish breeding populations. This study considered the following mating des...
详细信息
The optimal contribution selection method and the simulated annealing algorithm were used to maximize the genetic gain and reduce inbreeding in fish breeding populations. This study considered the following mating designs: full factorial (3 x 3);full factorial (6 x 6) and nested (6 males x 18 females). A quantitative trait based on a strictly additive and polygenic model was simulated. Two levels for the number of genotyped offspring (360 or 720) and three levels of heritability (0.1;0.3;0.5) were assumed. The best results in terms of Delta F and Delta G were obtained with the full factorial design (6 x 6) and considering a trait with a high heritability. The optimal family size was found at 20 fish per mating.
Rolling temperature is an important factor affecting mechanical properties of hot rotted strip significantly. Traditional techniques cannot meet higher precision control imperatives. In the present work, a novel knowl...
详细信息
ISBN:
(纸本)9780878494750
Rolling temperature is an important factor affecting mechanical properties of hot rotted strip significantly. Traditional techniques cannot meet higher precision control imperatives. In the present work, a novel knowledge-based system has been developed for the temperature prediction in hot strip mills. Neural network has been used for this purpose, which is an intelligent technique that can solve nonlinear problem of temperature control by learning from the samples. Furthermore, an annealing robust learning algorithm was presented to adjust the hidden node parameters as well as the weights of the adaptive neural networks. Simulations in a multi-object mode have been carried out to verify the effectivity of new neural optimization system. Calculation results confirm the feasibility of this approach and show a good agreement with experimental values obtained from a steel plant.
There are very important relations between transverse magnetic leakage and the ampere-turn balance for power transformer, which also enormously influence the axial short-circuit electromagnetic force acting on the tra...
详细信息
ISBN:
(纸本)9781479900688
There are very important relations between transverse magnetic leakage and the ampere-turn balance for power transformer, which also enormously influence the axial short-circuit electromagnetic force acting on the transformer windings and dynamic stability. In this paper, a novel method of ampereturn distribution is presented, and the annealing algorithm is used to optimize the distributions of ampere-turns. The model of 110kV power transformer is taken as an example for calculating. Results show that, after optimization, unbalanced ampere-turn is smaller than the former. The electromagnetic force, dynamic stress, and dynamic displacement on coil disks are simulated with the finite element software. Meanwhile, results demonstrate the electromagnetic force, dynamic stress and dynamic displacement are decreased to a certain extent.
暂无评论