The Experts below are selected from a list of 312 Experts worldwide ranked by ideXlab platform
Hui Tong Chua - One of the best experts on this subject based on the ideXlab platform.
-
two bed silica gel water adsorption chillers an effectual lumped Parameter Model
International Journal of Refrigeration-revue Internationale Du Froid, 2007Co-Authors: Xiaolin Wang, Hui Tong ChuaAbstract:This article develops an improved lump-Parameter design Model to investigate the water-circulation heat recovery scheme as applied to the two-bed silica gel–water adsorption chillers. We demonstrate that performance predictions stemming from this improved lump-Parameter formalism compare favorably with experimental results at various conditions, particularly at the industrial rated conditions. We find that the present lump-Parameter formalism adequately elucidates the water-circulation heat recovery scheme as does the distributed-Parameter formalism. In the studied working condition of a two-bed silica gel–water adsorption chiller, the differences in cooling capacities and coefficients of performance (or COP) by using the two different formalisms are typically less than 10%. This gives rise to a useful and rapid design tool for the industry.
Kristopher L. Kuhlman - One of the best experts on this subject based on the ideXlab platform.
-
Unsaturated hydraulic conductivity Models based on truncated lognormal pore-size distributions.
Ground water, 2014Co-Authors: Bwalya Malama, Kristopher L. KuhlmanAbstract:We develop a closed-form three-Parameter Model for unsaturated hydraulic conductivity associated with the Kosugi three-Parameter lognormal moisture retention Model. The Model derivation uses a slight modification to Mualem's theory, which is nearly exact for nonclay soils. Kosugi's three-Parameter lognormal moisture retention Model uses physically meaningful Parameters, but a corresponding closed-form relative hydraulic conductivity Model has never been developed. The Model is further extended to a four-Parameter Model by truncating the underlying pore-size distribution at physically permissible minimum and maximum pore radii. The proposed closed-form Models are fitted to well-known experimental data to illustrate their utility. They have the same physical basis as Kosugi's two-Parameter Model, but are more general.
-
Unsaturated Hydraulic Conductivity Models Based on Truncated Lognormal Pore-Size
2014Co-Authors: Distributions Bwalya Malama, Kristopher L. KuhlmanAbstract:We develop a closed-form three-Parameter Model for unsaturated hydraulic conductivity associated with the Kosugi three-Parameter lognormal moisture retention Model. The Model derivation uses a slight modification to Mualem’s theory, which is nearly exact for nonclay soils. Kosugi’s three-Parameter lognormal moisture retention Model uses physically meaningful Parameters, but a corresponding closed-form relative hydraulic conductivity Model has never been developed. The Model is further extended to a four-Parameter Model by truncating the underlying pore-size distribution at physically permissible minimum and maximum pore radii. The proposed closed-form Models are fitted to well-known experimental data to illustrate their utility. They have the same physical basis as Kosugi’s two-Parameter Model, but are more general.
-
Models for Unsaturated Hydraulic Conductivity Based on Truncated Lognormal Pore-size Distributions.
Journal of Hydrology, 2012Co-Authors: Bwalya Malama, Kristopher L. KuhlmanAbstract:Abstract We develop a closed-form three-Parameter Model for unsaturated hydraulic conductivity associated with a three-Parameter lognormal Model of moisture retention, which is based on lognormal grainsize distribution. The derivation of the Model is made possible by a slight modification to the theory of Mualem. We extend the three-Parameter lognormal distribution to a four-Parameter Model that also truncates the pore size distribution at a minimum pore radius. We then develop the corresponding four-Parameter Model for moisture retention and the associated closed-form expression for unsaturated hydraulic conductivity. The four-Parameter Model is fitted to experimental data, similar to the Models of Kosugi and van Genuchten. The proposed four-Parameter Model retains the physical basis of Kosugi’s Model, while improving fit to observed data especially when simultaneously fitting pressure-saturation and pressure-conductivity data.
Alex Gorodetsky - One of the best experts on this subject based on the ideXlab platform.
-
bayesian system id optimal management of Parameter Model and measurement uncertainty
Nonlinear Dynamics, 2020Co-Authors: Nicholas Galioto, Alex GorodetskyAbstract:System identification of dynamical systems is often posed as a least squares minimization problem. The aim of these optimization problems is typically to learn either propagators or the underlying vector fields from trajectories of data. In this paper, we study a first principles derivation of appropriate objective formulations for system identification based on probabilistic principles. We compare the resulting inference objective to those used by emerging data-driven methods based on dynamic mode decomposition (DMD) and system identification of nonlinear dynamics (SINDy). We show that these and related least squares formulations are specific cases of a more general objective function. We also show that the more general objective function yields more robust and reliable recovery in the presence of sparse data and noisy measurements. We attribute this success to an explicit accounting of imperfect Model forms, Parameter uncertainty, and measurement uncertainty. We study the computational complexity of an approximate marginal Markov Chain Monte Carlo method to solve the resulting inference problem and numerically compare our results on a number of canonical systems: linear pendulum, nonlinear pendulum, the Van der Pol oscillator, the Lorenz system, and a reaction–diffusion system. The results of these comparisons show that in cases where DMD and SINDy excel, the Bayesian approach performs equally well, and in cases where DMD and SINDy fail to produce reasonable results, the Bayesian approach remains robust and can still deliver reliable results.
-
bayesian system id optimal management of Parameter Model and measurement uncertainty
arXiv: Machine Learning, 2020Co-Authors: Nicholas Galioto, Alex GorodetskyAbstract:We evaluate the robustness of a probabilistic formulation of system identification (ID) to sparse, noisy, and indirect data. Specifically, we compare estimators of future system behavior derived from the Bayesian posterior of a learning problem to several commonly used least squares-based optimization objectives used in system ID. Our comparisons indicate that the log posterior has improved geometric properties compared with the objective function surfaces of traditional methods that include differentially constrained least squares and least squares reconstructions of discrete time steppers like dynamic mode decomposition (DMD). These properties allow it to be both more sensitive to new data and less affected by multiple minima --- overall yielding a more robust approach. Our theoretical results indicate that least squares and regularized least squares methods like dynamic mode decomposition and sparse identification of nonlinear dynamics (SINDy) can be derived from the probabilistic formulation by assuming noiseless measurements. We also analyze the computational complexity of a Gaussian filter-based approximate marginal Markov Chain Monte Carlo scheme that we use to obtain the Bayesian posterior for both linear and nonlinear problems. We then empirically demonstrate that obtaining the marginal posterior of the Parameter dynamics and making predictions by extracting optimal estimators (e.g., mean, median, mode) yields orders of magnitude improvement over the aforementioned approaches. We attribute this performance to the fact that the Bayesian approach captures Parameter, Model, and measurement uncertainties, whereas the other methods typically neglect at least one type of uncertainty.
Xiaolin Wang - One of the best experts on this subject based on the ideXlab platform.
-
two bed silica gel water adsorption chillers an effectual lumped Parameter Model
International Journal of Refrigeration-revue Internationale Du Froid, 2007Co-Authors: Xiaolin Wang, Hui Tong ChuaAbstract:This article develops an improved lump-Parameter design Model to investigate the water-circulation heat recovery scheme as applied to the two-bed silica gel–water adsorption chillers. We demonstrate that performance predictions stemming from this improved lump-Parameter formalism compare favorably with experimental results at various conditions, particularly at the industrial rated conditions. We find that the present lump-Parameter formalism adequately elucidates the water-circulation heat recovery scheme as does the distributed-Parameter formalism. In the studied working condition of a two-bed silica gel–water adsorption chiller, the differences in cooling capacities and coefficients of performance (or COP) by using the two different formalisms are typically less than 10%. This gives rise to a useful and rapid design tool for the industry.
-
Transient Modeling of a two-bed silica gel–water adsorption chiller
International Journal of Heat and Mass Transfer, 2004Co-Authors: Hui Chua, Weizong Wang, Christopher Yap, Xiaolin WangAbstract:This article presents a transient distributed-Parameter Model for a two-bed, silica gel–water adsorption chiller. Compared with our previous lumped-Parameter Model, we found better agreement between our Model prediction and experimental data. We discussed the important effect of heat recovery and the effect of extra system piping on the system performance. Time constants of sensors were also considered. We found that the chiller was able to maintain its cooling capacity over a fairly broad range of cycle times and the previous lumped-Parameter Model tended to under-predict the cooling capacity at long cycle times.
Nicholas Galioto - One of the best experts on this subject based on the ideXlab platform.
-
bayesian system id optimal management of Parameter Model and measurement uncertainty
Nonlinear Dynamics, 2020Co-Authors: Nicholas Galioto, Alex GorodetskyAbstract:System identification of dynamical systems is often posed as a least squares minimization problem. The aim of these optimization problems is typically to learn either propagators or the underlying vector fields from trajectories of data. In this paper, we study a first principles derivation of appropriate objective formulations for system identification based on probabilistic principles. We compare the resulting inference objective to those used by emerging data-driven methods based on dynamic mode decomposition (DMD) and system identification of nonlinear dynamics (SINDy). We show that these and related least squares formulations are specific cases of a more general objective function. We also show that the more general objective function yields more robust and reliable recovery in the presence of sparse data and noisy measurements. We attribute this success to an explicit accounting of imperfect Model forms, Parameter uncertainty, and measurement uncertainty. We study the computational complexity of an approximate marginal Markov Chain Monte Carlo method to solve the resulting inference problem and numerically compare our results on a number of canonical systems: linear pendulum, nonlinear pendulum, the Van der Pol oscillator, the Lorenz system, and a reaction–diffusion system. The results of these comparisons show that in cases where DMD and SINDy excel, the Bayesian approach performs equally well, and in cases where DMD and SINDy fail to produce reasonable results, the Bayesian approach remains robust and can still deliver reliable results.
-
bayesian system id optimal management of Parameter Model and measurement uncertainty
arXiv: Machine Learning, 2020Co-Authors: Nicholas Galioto, Alex GorodetskyAbstract:We evaluate the robustness of a probabilistic formulation of system identification (ID) to sparse, noisy, and indirect data. Specifically, we compare estimators of future system behavior derived from the Bayesian posterior of a learning problem to several commonly used least squares-based optimization objectives used in system ID. Our comparisons indicate that the log posterior has improved geometric properties compared with the objective function surfaces of traditional methods that include differentially constrained least squares and least squares reconstructions of discrete time steppers like dynamic mode decomposition (DMD). These properties allow it to be both more sensitive to new data and less affected by multiple minima --- overall yielding a more robust approach. Our theoretical results indicate that least squares and regularized least squares methods like dynamic mode decomposition and sparse identification of nonlinear dynamics (SINDy) can be derived from the probabilistic formulation by assuming noiseless measurements. We also analyze the computational complexity of a Gaussian filter-based approximate marginal Markov Chain Monte Carlo scheme that we use to obtain the Bayesian posterior for both linear and nonlinear problems. We then empirically demonstrate that obtaining the marginal posterior of the Parameter dynamics and making predictions by extracting optimal estimators (e.g., mean, median, mode) yields orders of magnitude improvement over the aforementioned approaches. We attribute this performance to the fact that the Bayesian approach captures Parameter, Model, and measurement uncertainties, whereas the other methods typically neglect at least one type of uncertainty.
