Nonlinear models have recently shown interesting properties for spectral unmixing. This paper studies a generalized bilinear model and a hierarchical bayesian algorithm for unmixing hyperspectral images. The proposed ...
详细信息
Nonlinear models have recently shown interesting properties for spectral unmixing. This paper studies a generalized bilinear model and a hierarchical bayesian algorithm for unmixing hyperspectral images. The proposed model is a generalization not only of the accepted linear mixing model but also of a bilinear model that has been recently introduced in the literature. Appropriate priors are chosen for its parameters to satisfy the positivity and sum-to-one constraints for the abundances. The joint posterior distribution of the unknown parameter vector is then derived. Unfortunately, this posterior is too complex to obtain analytical expressions of the standard bayesian estimators. As a consequence, a Metropolis-within-Gibbs algorithm is proposed, which allows samples distributed according to this posterior to be generated and to estimate the unknown model parameters. The performance of the resulting unmixing strategy is evaluated via simulations conducted on synthetic and real data.
We propose a bayesian based algorithm to recover sparse signals from compressed noisy measurements in the presence of a smooth background component. This problem is closely related to robust principal component analys...
详细信息
We propose a bayesian based algorithm to recover sparse signals from compressed noisy measurements in the presence of a smooth background component. This problem is closely related to robust principal component analysis and compressive sensing, and is found in a number of practical areas. The proposed algorithm adopts a hierarchical bayesian framework for modeling, and employs approximate inference to estimate the unknowns. Numerical examples demonstrate the effectiveness of the proposed algorithm and its advantage over the current state-of-the-art solutions.
This research work is aimed to carryout experimental investigation on copper slag incorporated self-compacting (SCC) concrete;here, copper slag is used as a replacement to fine aggregate in the range of 0-100%. The se...
详细信息
This research work is aimed to carryout experimental investigation on copper slag incorporated self-compacting (SCC) concrete;here, copper slag is used as a replacement to fine aggregate in the range of 0-100%. The self-compacting concrete is manufactured with powder matrix incorporating cement, flyash, metakaolin, and silicafume. The powder matrix is decided based on the objective of flow properties and strength properties. The fresh concrete properties of various mixes were studied and collected. Then, the concrete is casted as cubes and cylinders and tested for its strength behavior. The flow properties of copper slag-replaced mix were within stipulated guideline values by EFNARC, and early age strength attainment gets affected by the replacement of copper slag. The collected experimental results were designed to a data set categorizing as input and target. This data set is then used as a parameter in feed forward artificial neural network (ANN) and a predictive model is developed. This predictive model is then compared with the existing experimental values and tested for its performance. The results show that ANN provides a reliable predictive model for both flow and strength properties.
This paper examines the efficiency and capability of Dynet, a recurrent neural network model for the prediction of the damage evolution during hot non-uniform, non-isothermal forging on the basis of a limited number o...
详细信息
This paper examines the efficiency and capability of Dynet, a recurrent neural network model for the prediction of the damage evolution during hot non-uniform, non-isothermal forging on the basis of a limited number of damage snapshots during the process. A bayesian algorithm is introduced to optimise the hyperparameters related to the noise level and weight decay. In order to examine the capability of the model to capture the underlying trends when presented with sparse and noisy evidence, a synthetic relation between damage accumulation in a metal matrix composite and strain, strain rate and deformation temperature has been used to generate training data (evidence) of varying accuracy and sparseness. The results show that the bayesian algorithm performs very well, and that no significant overfitting is observed. In addition, this algorithm not only gives the expectation value of damage level, but also an estimate of its uncertainty. (c) 2005 Elsevier B.V. All rights reserved.
This study presents a novel strategy based on bayesian support vector regression for the estimation of the specific heat capacity of nitrides/ethylene glycol-based nanofluid. The nanoparticles considered are aluminium...
详细信息
This study presents a novel strategy based on bayesian support vector regression for the estimation of the specific heat capacity of nitrides/ethylene glycol-based nanofluid. The nanoparticles considered are aluminium nitride (AlN), silicon nitride (Si3N4) and titanium nitride (TiN). The proposed model was built using simple and easy-toobtain inputs such as the size of the nanoparticles (20, 30, 50, and 80 nm), the molar mass of the nanoparticles, mass fraction of nanoparticles (0.01 - 0.1) and the temperature (288.15 K, 298.15 K, and 308.15 K). Our suggested model showed better prediction accuracy over the analytical models for the estimation of specific heat capacity of nitrides/ethylene glycol nanofluids. Given the simplicity of the model inputs and the accuracy of the model, the approach presented provides a more reliable prediction of specific heat capacity of nitrides-ethylene glycol-based nanofluids than previous models.
This paper presents the modelling of the specific heat capacity (SHC) of CuO/water nanofluids using a support vector regression (SVR) and artificial neural network models (ANN). The models presented were developed fro...
详细信息
This paper presents the modelling of the specific heat capacity (SHC) of CuO/water nanofluids using a support vector regression (SVR) and artificial neural network models (ANN). The models presented were developed from the experimental data of SCH of CuO nanoparticles, the volume fractions of CuO nanoparticles and fluid temperature. The volume fraction of CuO nanoparticles considered ranges from 0.4 to 2% while the temperature range includes 293-338 K. The results obtained revealed that the SVR model exhibits slightly higher accuracy compared to the ANN model. However, both the SVR and ANN models clearly demonstrate better prediction performance for the SHC of CuO/water nanofluids compared to the existing theoretical models. The results obtained in this study proves that machine learning models provide a more accurate estimation of SHC of CuO/water nanofluids and they are recommended for heat transfer calculations due to their superior accuracy.
Carbon emission quota trading is an effective method for realizing carbon neutrality. Negotiation can help to effectively resolve the differences and conflicts among the trading parties in the carbon emission quota tr...
详细信息
Carbon emission quota trading is an effective method for realizing carbon neutrality. Negotiation can help to effectively resolve the differences and conflicts among the trading parties in the carbon emission quota trading process. For researching the negotiation mechanism of carbon emission quota trade, we construct the agent negotiation model that introduces the risk appetite of negotiator and bayesian algorithm. And then, the nego-tiation protocol is analyzed, including offer tactics, negotiation unity and termination condition. The experi-mental studies are given to simulate the negotiation of carbon emission quota trading process for demonstrating the validity of the negotiation model and tactics. This study demonstrates the impact of risk appetite on nego-tiation in the carbon emission quota trading. And can help to improve negotiation efficiency, achieve win-win and accurate matching among the trading parties of carbon emission quota.
There is a certain subjectivity in the teaching evaluation process, which leads to a low accuracy of the intelligent scoring system. In order to promote the intelligent development of teaching evaluation, based on mac...
详细信息
There is a certain subjectivity in the teaching evaluation process, which leads to a low accuracy of the intelligent scoring system. In order to promote the intelligent development of teaching evaluation, based on machine learning, this study briefly introduces the background and current status of teaching evaluation, and describes in detail the relevant algorithm principles of data analysis and modeling using data mining technology and machine learning methods. Moreover, this study describes the establishment process of the traditional classroom teaching evaluation system and uses the classification algorithm in machine learning in the construction of evaluation models to further improve the scientificity and feasibility of teaching evaluation. In addition, in this study, empirical algorithm is used as the basic algorithm to evaluate teaching quality, and the topic word distribution obtained by joint model training is used as the original knowledge. Finally, this research analyzes the performance of this research system through a control experiment. The research results show that the scores of the research model are close to the standard manual scores and can provide a theoretical reference for subsequent related research.
This article proposes a hybrid method (HM) to improve the accuracy of short-term individual residential load forecasting. The HM includes an ensemble model (EM), deep ensemble model (DEM), and thermal dynamic model ex...
详细信息
This article proposes a hybrid method (HM) to improve the accuracy of short-term individual residential load forecasting. The HM includes an ensemble model (EM), deep ensemble model (DEM), and thermal dynamic model expressed by resistance-capacitance (RC). The EM consists of three predictors of support vector machine (SVM), back propagation neural network (BPNN), and generalized regression neural network (GRNN). The genetic algorithm (GA) is used to optimize SVM and BPNN to enhance their performance. The DEM includes multiple bi-directional long-short term memory (Bi-LSTM) networks. The bayesian algorithm (BA) is used to optimize the hyperparameters of the Bi-LSTM. The outputs of individual predictors are aggregated using an optimal trimmed algorithm. At first, the total load is separated into the heater and air conditioning (HAC), and non-HAC loads. Then, the RC model is presented to predict the indoor temperature, which integrates outdoor weather and less HAC historical data as the input of the EM to forecast the HAC load. After that, non-HAC loads are further divided into electric lighting and other loads. A daylight equation is used to calculate the illuminance, which is combined with less lighting historical data as the input of DEM to predict electric lights usage. Then, other loads are captured by DEM through less historical data. Finally, the total load is obtained by combining the predicted HAC and non-HAC loads. The datasets from the UMass Smart Microgrid and Flexhouse projects are used to test the proposed method. The comparison with existing models proves that the presented model can provide accurate short-term individual load forecasting.
Tensioned fabric membrane structures, in which a coated fabric material is used as both cladding and supporting structure, are classified as unconventional structures and demand a dedicated design method. In this pape...
详细信息
Tensioned fabric membrane structures, in which a coated fabric material is used as both cladding and supporting structure, are classified as unconventional structures and demand a dedicated design method. In this paper, we propose a method to combine different stages in design and analysis of these structures within the shape optimization framework. The proposed method facilitates the incorporation of various sources of nonlinearities in the analysis of tensioned membrane structures. Especially, the amount of fabric panels compensation when using simple material model for the coated fabric, i.e., linear orthotropic elasticity, as is widely used in practice, is presented. The novelty of the proposed method lies on the postulation of a stress-free 3-D intermediate configuration, which is built from flat fabric panels. The goal is to find an optimum shape of this 3-D intermediate configuration such that the stress distribution in the resulting structure, which is formed by deforming the 3-D intermediate configuration, remains at desired levels. A nonlinear finite element analysis is employed to identify these stress fields in the resulting structures. In this analysis, material and geometrical nonlinearities are taken into account. Moreover, the nonlinear kinematical interaction between the boundary cables and the membrane is also considered. Hence the stress fields on the resulting structures are expected to be reliable and close to the physical ones. In this work, different gradientless optimization algorithms, viz., genetic algorithm, pattern search, and bayesian, are considered. The obtained results show that the stress levels in the resulting structures obtained from the proposed method are close to the desired ones, and the bayesian optimization algorithm has the best performance among the considered algorithms. (C) 2017 Elsevier Ltd. All rights reserved.
暂无评论