Prediction of programmed-temperature retention values of naphthas by wavelet neural networks

Zhang, Xiaoyun; Qi, Jianhua; Zhang, Ruisheng; Liu, Mancang; Hu, Zhide; Xue, Huifeng; Fan, Bo

doi:10.1016/s0097-8485(00)00074-7

Cited by 61 publications

(21 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although RBF is also local function, but it does not have the spatial-spectral zooming property of the wavelet function, and therefore it cannot represent the local spatial spectral characteristics of the function. WNN shows surprising effectiveness in solving the conventional problems of poor convergence or even divergence encountered in other kinds of neural networks [36].…”

Section: Wavelet Neural Networkmentioning

confidence: 99%

Ensemble ANNs-PSO-GA Approach for Day-ahead Stock E-exchange Prices Forecasting

Xiao¹,

Xiao²,

Lu³

et al. 2014

IJCIS

View full text Add to dashboard Cite

Stock e-exchange prices forecasting is an important financial problem that is receiving increasing attention. This study proposes a novel three-stage nonlinear ensemble model. In the proposed model, three different types of neural-network based models, i.e. Elman network, generalized regression neural network (GRNN) and wavelet neural network (WNN) are constructed by three non-overlapping training sets and are further optimized by improved particle swarm optimization (IPSO). Finally, a neural-network-based nonlinear meta-model is generated by learning three neural-network based models through support vector machines (SVM) neural network. The superiority of the proposed approach lies in its flexibility to account for potentially complex nonlinear relationships. Three daily stock indices time series are used for validating the forecasting model. Empirical results suggest the ensemble ANNs-PSO-GA approach can significantly improve the prediction performance over other individual models and linear combination models listed in this study.

show abstract

Section: Wavelet Neural Networkmentioning

confidence: 99%

Ensemble ANNs-PSO-GA Approach for Day-ahead Stock E-exchange Prices Forecasting

Xiao¹,

Xiao²,

Lu³

et al. 2014

IJCIS

View full text Add to dashboard Cite

show abstract

“…These characteristics make wavelets an active subject with many exciting applications, not only in pure mathematics, but also in acoustics, image compression, turbulence, human vision, radar, earthquake prediction, fluid mechanics and chemical analysis. 26 A series of wavelets is generated by stretching and shifting the wavelet over the data. The shift b is called a translation and the stretching or widening of the basis wavelet with a factor a is called a dilation.…”

Section: Wavelet and Wavelet Neural Networkmentioning

confidence: 99%

Quantitative Structure-Property Relationship Study of the Solvent Polarity Using Wavelet Neural Networks

2007

View full text Add to dashboard Cite

Quantitative structure-activity and structure-property relationship (QSAR/QSPR) studies are unquestionably of great importance in modern chemistry and biochemistry. The concept of QSAR/QSPR is to transform searches for compounds with desired properties using chemical intuition and experience into a mathematically quantified and computerized form. 1 Solvent polaritySimplistically, "solutions" can be defined as homogeneous liquid phases consisting of more than one substance in variable ratios, where one of the substances, the so-called "solvent" (which could itself be a mixture), is treated differently from the other substances, which are called "solutes". Usually, the solute(s) is/are the minor component(s) and the solvent is the component in excess. The term "polarity" is usually related to the capacity of a solvent for solvating dissolved charged or dipolar species.2 Solvent polarity defined as the overall solvation capability (or solvation power) for reactants and activated complexes as well as for molecules in the ground and excited states, which in turn depends on the action of all possible, specific and nonspecific, intermolecular forces between solvent and solute molecules, including Coulomb interactions between ions, directional interactions between dipoles, and inductive, dispersion, hydrogen-bonding, and charge-transfer forces, as well as solvophobic interactions. Only those interactions leading to definite chemical alterations of the solute molecules through protonation, oxidation, reduction, complex formation, or other chemical processes are excluded. 3 A number of general empirical solvent scales have been reported. [3][4][5] In earlier reports, a unified solvation model was offered to treat both specific and non specific effects. 5,6-10 In 1992, Drago 5 developed the "unified solvent polarity", also called the "S′ scale" by using a least-squares minimization program 11 to fit a series of physico-chemical properties (χ) for systems where specific interactions with the solvents were excluded to the equation:In this equation, Δχ is the solvent-dependent physicochemical property, P is the susceptibility of the solute probe to polarity, S′ is the solvent polarity derived from experimental observations, EA and EB are the electrostatic specific interaction parameters of the acid and base, CA and CB are the covalent parameters of the acid and base and W is a constant that is the value of Δχ when EA = CA = P = 0. This model separates solvent effects into non specific (PS′) and specific (EAEB + CACB) interactions. The former arises from non specific electrostatic forces involving bulk and internal solvent permittivity influences on charged ions, non polar and dipolar solutes. The latter comprises electron-pair, donor-acceptor interactions including hydrogenbonding and π-π* charge transfer. 12Katritzky and co-workers 13 developed a three-parameter QSPR model (N = 25 solvents, R 2 = 0.936), and Mu and coworkers 14 developed a two-parameter MQSPR (model-based QSPR) model for a unified solvent polarity scale (S′...

show abstract

“…Wavelet function is a local function and influences the networks' output only in some local ranges. The wavelet neural network shows surprising effectiveness in solving the conventional problems of poor convergence or even divergence encountered in other kinds of neural networks [12]. The WNN consists of three layers: input layer, hidden layer and output layer.…”

Section: Basic Concepts Of Wavelet Neural Networkmentioning

confidence: 99%

A New Hybrid Approach for Analysis of Factors Affecting Crude Oil Price

Wang

Zhang

et al. 2007

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper, a new hybrid approach is presented to analyze factors affecting crude oil price using rough set and wavelet neural network. Related factors that affect crude oil price are found using text mining technique and Brent oil price is chosen as the decision price because it plays an important role in world crude oil markets. The relevant subsets of the factors are discovered by rough set module and the main factors are got, and then the important degrees of these are measured using wavelet neural network. Based on the novel hybrid approach, the predictability of crude oil price is discussed.

show abstract

Prediction of programmed-temperature retention values of naphthas by wavelet neural networks

Cited by 61 publications

References 21 publications

Ensemble ANNs-PSO-GA Approach for Day-ahead Stock E-exchange Prices Forecasting

Ensemble ANNs-PSO-GA Approach for Day-ahead Stock E-exchange Prices Forecasting

Quantitative Structure-Property Relationship Study of the Solvent Polarity Using Wavelet Neural Networks

A New Hybrid Approach for Analysis of Factors Affecting Crude Oil Price

Contact Info

Product

Resources

About