Estimation of thermophysical properties of dimethyl ether as a commercial refrigerant based on artificial neural networks

Moghadassi, Abdolreza; Nikkholgh, M. R.; Parvizian, Fahime; Hosseini, S.M.

doi:10.1016/j.eswa.2010.04.065

Cited by 32 publications

(8 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It achieves regression by building a model of the data-generating process for the network to generalize and predict outputs from inputs that are not previously seen. In this study, back-prorogation (BP) neural networks-based regression were utilized in that it can handle non-linear relationships among data even when there are conflicting relationships between the input variables and the response variables (Moghadassi et al, 2010). The optimal number of hidden layers and neurons in the BP neural networks was determined through the leave-one-out cross validation process that yielded the smallest RMSE value.…”

Section: Regression Techniquesmentioning

confidence: 99%

Hyperspectral Leaf Reflectance as Proxy for Photosynthetic Capacities: An Ensemble Approach Based on Multiple Machine Learning Algorithms

Meacham-Hensold

Guan

et al. 2019

Front. Plant Sci.

View full text Add to dashboard Cite

Global agriculture production is challenged by increasing demands from rising population and a changing climate, which may be alleviated through development of genetically improved crop cultivars. Research into increasing photosynthetic energy conversion efficiency has proposed many strategies to improve production but have yet to yield real-world solutions, largely because of a phenotyping bottleneck. Partial least squares regression (PLSR) is a statistical technique that is increasingly used to relate hyperspectral reflectance to key photosynthetic capacities associated with carbon uptake (maximum carboxylation rate of Rubisco, V c,max ) and conversion of light energy (maximum electron transport rate supporting RuBP regeneration, J max ) to alleviate this bottleneck. However, its performance varies significantly across different plant species, regions, and growth environments. Thus, to cope with the heterogeneous performances of PLSR, this study aims to develop a new approach to estimate photosynthetic capacities. A framework was developed that combines six machine learning algorithms, including artificial neural network (ANN), support vector machine (SVM), least absolute shrinkage and selection operator (LASSO), random forest (RF), Gaussian process (GP), and PLSR to optimize high-throughput analysis of the two photosynthetic variables. Six tobacco genotypes, including both transgenic and wild-type lines, with a range of photosynthetic capacities were used to test the framework. Leaf reflectance spectra were measured from 400 to 2500 nm using a high-spectral-resolution spectroradiometer. Corresponding photosynthesis vs. intercellular CO 2 concentration response curves were measured for each leaf using a leaf gas-exchange system. Results suggested that the mean R 2 value of the six regression techniques for predicting V c,max ( J max ) ranged from 0.60 (0.45) to 0.65 (0.56) with the mean RMSE value varying from 47.1 (40.1) to 54.0 (44.7) μmol m -2 s -1 . Regression stacking for V c,max ( J max ) performed better than the individual regression techniques with increases in R 2 of 0.1 (0.08) and decreases in RMSE by 4.1 (6.6) μmol m -2 s -1 , equal to 8% (15%) reduction in RMSE . Better predictive performance of the regression stacking is likely attributed to the varying coefficients (or weights) in the level-2 model (the LASSO model) and the diverse ability of each individual regression technique to utilize spectral information for the best modeling performance. Further refinements can be made to apply this stacked regression t...

show abstract

Section: Regression Techniquesmentioning

confidence: 99%

Hyperspectral Leaf Reflectance as Proxy for Photosynthetic Capacities: An Ensemble Approach Based on Multiple Machine Learning Algorithms

Meacham-Hensold

Guan

et al. 2019

Front. Plant Sci.

View full text Add to dashboard Cite

show abstract

“…The properties obtained from these models were validated against most reliable NIST-REFPROP data base. Investigation on prediction of liquid density and vapour pressure of refrigerant RE170 (Dimethylether) was done based on artificial neural networks using back-propagation algorithm [8]. Prediction of above properties exhibited good agreement with the experimental data.…”

Section: Introductionmentioning

confidence: 86%

“…where, C, S, H, X and O are number of carbons, sulfur, hydrogen, halogen and oxygen atoms presented in the given refrigerant, respectively. By using equations (6)(7)(8), values of flammability limits of refrigerant RE170 are calculated as 3.34 vol% and 25.60 vol%, respectively.…”

Section: Computation Of Flammability Properties Of Refrigerant Re170mentioning

confidence: 99%

Computation of Ecological Properties, Flammability Properties and Thermodynamic Properties of Sustainable Refrigerant Dimethylether (RE170) Using Martin Hou Equation of State (MHEOS)

Shaik¹,

Babu²

2019

IJHT

View full text Add to dashboard Cite

The objective of present work is to investigate the thermodynamic properties, ecological properties and flammability properties of a refrigerant Dimethylether (RE170). Ecological properties such as ozone depletion potential (ODP) and global warming potential (GWP) were reported and validated with literature. Flammability properties such as upper flammability limit (UFL), lower flammability limit (LFL) and refrigerant flammability (RF) number were reported and validated with literature. Martin-Hou equation of state (MHEOS) is used while computing the thermodynamic properties of RE170. In this investigation, a MATLAB code was developed in order to compute the properties of RE170 over the wide temperature range about 131.6 K to 398.15 K and pressure up to 51.3 bar. The computed thermodynamic properties of RE170 were validated against REFPROP database, since its thermodynamic properties were not available in the existing literature. Therefore, REFRPROP can be considered as a reliable source as that of ASHRAE. Results showed that refrigerant RE170 has zero ODP and very low GWP (equals to one). Similarly RE170 has LFL of 3.34 vol%, UFL of 25.60 vol% and RF number of 51.05 kJ/g. Results of thermodynamic properties of RE170 revealed that the mean absolute deviation (MAD) between computed liquid and vapour phase enthalpy values and REFPROP values was 1.01 % and 0.32 %, respectively. Similarly MAD between computed liquid and vapour phase entropy values and REFPROP values was 1.05 % and 0.25 %. MAD of saturation pressure was 0.071 %. MAD between computed vapour specific volume, liquid density values and REFPROP values was 1.49 % and 0.25 %. Overall, computed thermodynamic, ecological and flammability properties of RE170 shows good agreement with literature.

show abstract

“…В схожей работе Мохебби и др. [25] предсказали плотность в области насыщенной жидкости для 19 чистых и 6 смесей хладагентов. В их работе, оптимальное количество нейронов в скрытом слое было найдено с помощью генетического алгоритма.…”

Section: использование нейронных се-тей для расчета термодинамичunclassified

Нейромережеве Прогнозування Властивостей Холодоагентів: Огляд

Artemenko¹

2015

Refrigeration Engineering and Technology

View full text Add to dashboard Cite

В работе проведен обзор статей, связанных с моделированием свойств хладагентов с помощью искусственных нейронных сетей, представляющих альтернативу прогнозированию термодинамических свойств веществ на основе аппарата термодинамики. В статье обсуждается эффективность, адекватность и ограничения применения этого подхода. Ключевые слова: Искусственные нейронные сети-Свойства хладагентов-Фазовое равновесие С.В. Артеменко Одеська національна академія харчових технологій, навчально-науковий інститут холоду, кріотехнологій та екоенергетики ім. В.С. Мартиновського, вул. Дворянська, 1/3, Одеса, 65082 НЕЙРОМЕРЕЖЕВЕ ПРОГНОЗУВАННЯ ВЛАСТИВОСТЕЙ ХОЛОДОАГЕНТІВ: ОГЛЯД В роботі проведено огляд статей, пов'язаних з моделюванням властивостей холодоагентів за допомогою штучних нейронних мереж, що представляють альтернативу прогнозуванню термодинамічних властивостей речовин на основі апарату термодинаміки. У статті обговорюється ефективність, адекватність і обмеження застосування цього підходу. Ключові слова: Штучні нейронні мережі-Властивості холодоагентів-Фазова рівновага

show abstract

Estimation of thermophysical properties of dimethyl ether as a commercial refrigerant based on artificial neural networks

Cited by 32 publications

References 26 publications

Hyperspectral Leaf Reflectance as Proxy for Photosynthetic Capacities: An Ensemble Approach Based on Multiple Machine Learning Algorithms

Hyperspectral Leaf Reflectance as Proxy for Photosynthetic Capacities: An Ensemble Approach Based on Multiple Machine Learning Algorithms

Computation of Ecological Properties, Flammability Properties and Thermodynamic Properties of Sustainable Refrigerant Dimethylether (RE170) Using Martin Hou Equation of State (MHEOS)

Нейромережеве Прогнозування Властивостей Холодоагентів: Огляд

Contact Info

Product

Resources

About