Mineral bioflotation optimization: Comparison between artificial neural networks and response surface methodology

Pereira, Anderson; Olivera, Carlos Alberto Castañeda; Merma, Antonio Gutiérrez; Hacha, Ronald Rojas; Santos, Brunno F.; Torem, Maurício Leonardo

doi:10.1016/j.mineng.2021.106983

Cited by 12 publications

(3 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The training settings of the BPNN model in this study are summarized as follows: number of input nodes: 17, number of hidden neurons: from 6 to 16, number of output nodes: 1, number of epochs: 1000. As stated in the literature [81], unlike changing the number of hidden nodes, changing the activation function does not have a significant effect on the model performance and results with similar MSE and R. The tangent sigmoid and linear activation functions were used in the hidden and output layers, respectively.…”

Section: Sulfur Removal Estimation Based On Mlrmentioning

confidence: 99%

“…It is important to note that sometimes BPNN does not represent the most cost-effective solution. In some cases, the interpretability of the network and weights may be difficult, the determination of the optimum structure and parameters may be troublesome, and the convergence of the training algorithm may be endless [81]. In these situations, MLR models can have advantages in terms of accuracy, variability, and model creation; therefore, its choice is preferred [83].…”

Section: Comparison Of Mlr K-means Bpnn and Cnn Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Prediction of Sulfur Removal from Iron Concentrate Using Column Flotation Froth Features: Comparison of k-Means Clustering, Regression, Backpropagation Neural Network, and Convolutional Neural Network

2022

View full text Add to dashboard Cite

Froth feature extraction plays a significant role in the monitoring and control of the flotation process. Image-based soft sensors have received a great deal of interest in the flotation process due to their low-cost and non-intrusive properties. This study proposes data-driven soft sensor models based on froth images to predict the key performance indicators of the flotation process. The ability of multiple linear regression (MLR), the backpropagation neural network (BPNN), the k-means clustering algorithm, and the convolutional neural network (CNN) to predict the amount of sulfur removal from iron ore concentrate in the column flotation process was examined. A total of 99 experimental results were used to develop the predictive models. Extracted froth features including color, bubble shape and size, texture, stability, and velocity were used to train the traditional predictive models, whereas in the CNN model the froth images were directly fed into the model. The results comparison indicated that the three-layered feedforward NN model (17-10-1 topology) and CNN model provided better predictions than the MLR and k-means algorithm. The BPNN model displayed a correlation coefficient of 0.97 and a root mean square error of 4.84% between the actual data and network output for both training and the testing datasets. The error percentages of the CNN, BPNN, MLR and k-means models were 10, 11, 15 and 18%, respectively. This study can become a key technical support for the application of intelligent models in the control of the operational variables for the flotation process used to desulfurize iron concentrate.

show abstract

Section: Sulfur Removal Estimation Based On Mlrmentioning

confidence: 99%

Section: Comparison Of Mlr K-means Bpnn and Cnn Modelsmentioning

confidence: 99%

Prediction of Sulfur Removal from Iron Concentrate Using Column Flotation Froth Features: Comparison of k-Means Clustering, Regression, Backpropagation Neural Network, and Convolutional Neural Network

2022

View full text Add to dashboard Cite

show abstract

“…These connections play a crucial role in adjusting weights and biases during network training. Once the training process is complete, the network becomes capable of predictive simulation within the range of the inputs provided in the training data [16]. Carvalho et al [17] developed artificial neural networks and observed higher reliability than the model with a linear polynomial equation incorporating rectangular interaction terms.…”

Section: Introductionmentioning

confidence: 99%

Modelagem robusta da formação de goma em misturas brasileiras de etanol-gasolina com base em planejamento de experimentos e abordagens de redes neurais artificiais

PRADELLE,

CARVALHO,

SANTOS

et al. 2023

Blucher Engineering Proceedings

View full text Add to dashboard Cite

This work aims to predict gum formation, based on experimental data from controlled laboratory tests. The tested conditions follow design of experiments approaches and allow to measure either unwashed or washed gum contents for different ethanol concentration, storage temperature and aging period in homologation gasoline. The response surfaces arising from the obtained mathematical models are compared with models developed with Artificial Neural Networks (ANN) available in the literature. In addition to its ability to generalize the impact of the factors, the ANN models showed the best performance. It is important to highlight that the catalytic effect of ethanol in gum formation that was observed by visualizing the available data is predicted by the ANN model, but is not statistically proven in the polynomial model. Thus, it is possible to verify a catalytic effect for low concentrations of ethanol (around 20 vol%) after aging. Without aging or after storage at low temperature (approximately 20ºC), a simple dilution effect is observed. Such conclusions provide a robust unified interpretation to justify the discrepancy between some authors from the literature. RESUMOEste trabalho tem como objetivo predizer a formação da goma, com base em dados experimentais obtidos em testes laboratoriais controlados. As condições testadas seguem abordagens de planejamento de experimentos e permitem medir os teores de goma não lavada ou lavada para diferentes níveis experimentais para a concentração de etanol, temperatura e período de envelhecimento em gasolina de homologação. As superfícies de resposta decorrentes dos modelos matemáticos obtidos são comparadas com modelos obtidos com Redes Neurais Artificiais (RNA) disponíveis na literatura. Além de sua capacidade de generalizar o impacto dos fatores, os modelos de RNA apresentaram o melhor desempenho,. É importante destacar que o efeito catalítico do etanol na formação da goma que foi especulado pela visualização dos dados disponíveis é previsto pelo modelo RNA, mas não está comprovado estatisticamente no modelo polinomial. Assim, é possível verificar um efeito catalítico para baixas concentrações de etanol (em torno de 20 vol%) após o envelhecimento.Sem envelhecimento ou após armazenamento a baixa temperatura (aproximadamente 20ºC), observa-se um simples efeito de diluição. Tais conclusões fornecem uma interpretação unificada e robusta para justificar a discrepância entre alguns autores da literatura.However, in spark-ignition engines, fossil gasoline, a complex blend of paraffinic, naphthenic, olefinic, and aromatic hydrocarbons, continues to be the primary fuel. It also contains a low concentration of oxygenates and traces of sulfur, nitrogen, and metals, which introduce instability to

show abstract

Polynomial and ANN models applied to the formation of gums in Brazilian ethanol–gasoline blends—impact of gasoline composition, ethanol concentration, storage temperature, and aging duration

Carvalho

Santos

Martins

et al. 2023

Neural Comput & Applic

View full text Add to dashboard Cite

Mineral bioflotation optimization: Comparison between artificial neural networks and response surface methodology

Cited by 12 publications

References 29 publications

Prediction of Sulfur Removal from Iron Concentrate Using Column Flotation Froth Features: Comparison of k-Means Clustering, Regression, Backpropagation Neural Network, and Convolutional Neural Network

Prediction of Sulfur Removal from Iron Concentrate Using Column Flotation Froth Features: Comparison of k-Means Clustering, Regression, Backpropagation Neural Network, and Convolutional Neural Network

Modelagem robusta da formação de goma em misturas brasileiras de etanol-gasolina com base em planejamento de experimentos e abordagens de redes neurais artificiais

Polynomial and ANN models applied to the formation of gums in Brazilian ethanol–gasoline blends—impact of gasoline composition, ethanol concentration, storage temperature, and aging duration

Contact Info

Product

Resources

About