A hybrid statistical approach for modeling and optimization of RON: A comparative study and combined application of response surface methodology (RSM) and artificial neural network (ANN) based on design of experiment (DOE)

Elfghi, Fawzi M.

doi:10.1016/j.cherd.2016.05.023

Cited by 73 publications

(37 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One of the ways to optimize the number of experiments is the statistical design of the experiment (DOE). From numerous procedures of DOE [52,53], in this work, the central composite design (CCD) was selected and constructed for three factors. The calibration samples (training set) contained constituent patterns: the full factorial points (values − 1 or + 1 on each axis), central point and its repetitions (value 0 on each axis), and star points (at some value − γ or + γ on each axis).…”

Section: Design Of the Experiments Using Central Composite Designmentioning

confidence: 99%

Multivariate approach in voltammetric identification and simultaneous determination of eugenol, carvacrol, and thymol on boron-doped diamond electrode

Kowalcze

Jakubowska

2019

Monatsh Chem

View full text Add to dashboard Cite

The paper presents multivariate partial least-squares calibration model defined for the identification and simultaneous determination of phenolic terpenoids-eugenol, carvacrol, and thymol-by means of the differential pulse voltammetry. For the first time, using boron-doped diamond electrode signal, electrochemical distinction of carvacrol and thymol, despite the short distance between the peaks (15 mV), was described. The analytes' concentrations applied in the calibration model were selected using the 3-factorial central composite design. The multivariate partial least-squares model, with optimized data pre-processing, was used for determination of eugenol, carvacrol, and thymol in the concentration ranges of 0.77-13.86, 0.58-10.40, and 0.73-7.32 mg/dm 3 , respectively. The quality of the method, expressed by the root-mean-squared error of calibration was 0.55, 0.20, and 0.26 mg/dm 3 for the specified analytes. The recovery in the analysis of the pure signals was 98.92%, 101.62%, and 101.46%, what confirmed the possibility of clearly distinguishing the thymol and carvacrol peaks. In the spiked honey samples, the accuracy of eugenol, carvacrol, and thymol determination, expressed by the root-meansquared error of prediction, was in the range 0.20-0.83 mg/dm 3 , and recovery in the range 80.9-111.9%. Application of the presented model not only enables electrochemical identification and simultaneous determination of eugenol, carvacrol, and thymol, but also allows rapid, accurate, and low-cost verification of authenticity of the commercial products.

show abstract

Section: Design Of the Experiments Using Central Composite Designmentioning

confidence: 99%

Multivariate approach in voltammetric identification and simultaneous determination of eugenol, carvacrol, and thymol on boron-doped diamond electrode

Kowalcze

Jakubowska

2019

Monatsh Chem

View full text Add to dashboard Cite

show abstract

“…The criterion used for the selection of the number of hidden neurons is the same as that assumed by Abdul Rahman et al [11] and Moradi et al [16]. They concluded that the ANN backpropagation model that showed the best performance to explain the relationship between the input and output variables of the reactions studied in their respective works included an intermediate number of neurons in the hidden layer.…”

Section: Resultsmentioning

confidence: 99%

“…So, the simplest neural structure, ANN1 (configuration 3-35-1, Fig. 1-a), was selected [9,11,15]. Consequently, it was chosen for evaluation, modifying the number of neurons in its hidden level.…”

Section: Resultsmentioning

confidence: 99%

“…They can estimate the practical relationships between the independent variables (input) of the process and the dependent ones (output), using data which were obtained experimentally. There is no criterion that can be applied to its constitution [11]. However, there are ANN architectures with more appropriate characteristics according to the problem to be solved.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

r Comparison of Neural Networks. An Estimation Model in Yield of Monoglycerides from Biodiesel by-Product

Álvarez¹,

Bálsamo²,

Modesti³

et al. 2019

JESTR

View full text Add to dashboard Cite

Biodiesel is generally manufactured by transesterification, obtaining glycerol as a by-product. The transesterification of methyl stearate selectively produced monoglycerides, for glycerol valuation. Mixed oxides containing lithium catalysed the reaction. The purpose of this work was to develop and compare mathematical models obtained through artificial neural networks (ANN), capable for characterising the relationship between the mole percent conversion of methyl stearate and the yield of the products mono-, di-and triglycerides. The lowest mean squared error (MSE), the highest correlation coefficient (R), similarity in the evolution of validation and simulation errors and absence of data overlearning were considered to select the best model. Three ANNs with backpropagation structures were compared. They evidenced high correspondence between the estimated product yield values and the interpolated experimental ones. The ANN containing 35 neurons with sigmoid transfer function in the hidden layer and a linear neuron in the output one was the simplest. Consequently, the 5, 15 and 60 neurons were also explored in the hidden layer. The ANN structured with an intermediate number of neurons (35) achieved the most adequate MSE, considering mono-and diglyceride products (0.011193, 0.000489). The development of these models contributes to the dynamic estimation of the process.

show abstract

“…Las redes neuronales artificiales se componen de elementos unitarios (neuronas) que operan conjuntamente y están inspiradas en el sistema biológico. La participación de cada neurona en la red se encuentra ponderada por medio de asignaciones escalares que se valoran en fase de entrenamiento (Elfghi, 2016). La lógica difusa, en tanto, posibilita la captura de información expresada en lenguaje natural, por medio de datos similares a los que es capaz de evaluar el cerebro humano, permitiendo una representación simple de los procesos en términos de reglas If -Then (Si tal cosa -entonces tal otra) (Adelkhani et al, 2013;Kurtener et al, 2016).…”

Section: Introductionunclassified

Sistema Adaptativo de Inferencias Difusas para la Caracterización de la Calidad Sensorial de Aceitunas Negras Naturales

Álvarez

Modesti

2018

Inf. tecnol.

View full text Add to dashboard Cite

ResumenEl objetivo del trabajo es aplicar la lógica difusa en la evaluación sensorial de aceitunas negras naturales, mediante un modelo que vincule la valoración de jueces sobre el color, textura y sabor, con la calidad sensorial del alimento. Se realizaron pruebas de diagnóstico por atributos y valoración general con 30 jueces. El modelo se desarrolló mediante un sistema adaptativo de inferencias difusas, ANFIS. Como entrada se tomó la calificación de los jueces sobre cada atributo y como salida, el valor asignado en la aceptación general. Se definió la ponderación de los atributos evaluados sobre el perfil general de la muestra, determinando tres funciones de membresía de campana generalizada. La red creada demostró desempeño aceptable, aún con un reducido número de neuronas empleados para su desarrollo. Palabras clave: sistema adaptativo; inferencias difusas, ANFIS; calidad sensorial; aceitunas negras AbstractThe objective of the work was the application of diffuse logic in the sensorial evaluation of natural black olives, through the development of a model that links the valuation made by consumer judges in relation to color, texture and flavor attributes, with the food general acceptance. Sensory evaluation was developed with 30 consumer judges. The model was generated using Adaptive Neuro-Fuzzy Inference System, ANFIS. The qualification of the judges for each attribute was considered as input data. The values assigned by them in the general acceptance tests were considered as the output matrix. The weighting of each of the attributes evaluated on the general profile of the sample was defined, which was delimited by means of three generalized bell-type membership functions. The ANFIS network created showed an acceptable performance, even the reduced number of neurons used for its development.

show abstract

A hybrid statistical approach for modeling and optimization of RON: A comparative study and combined application of response surface methodology (RSM) and artificial neural network (ANN) based on design of experiment (DOE)

Cited by 73 publications

References 37 publications

Multivariate approach in voltammetric identification and simultaneous determination of eugenol, carvacrol, and thymol on boron-doped diamond electrode

Multivariate approach in voltammetric identification and simultaneous determination of eugenol, carvacrol, and thymol on boron-doped diamond electrode

r Comparison of Neural Networks. An Estimation Model in Yield of Monoglycerides from Biodiesel by-Product

Sistema Adaptativo de Inferencias Difusas para la Caracterización de la Calidad Sensorial de Aceitunas Negras Naturales

Contact Info

Product

Resources

About