Artificial neural network-based modelling of optimized experimental study of xylanase production by <i>Penicillium citrinum</i> xym2

Kumar, Gaurav; Saha, Shyama Prasad; Ghosh, Shilpi; Mondal, Pranab Kumar

doi:10.1177/09544089211064153

Cited by 8 publications

(7 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although in this experimental study, MSE is used as the main performance criterion, as is also used in other studies [31,32], additional performance evaluators can also be used for this purpose. Together with MSE there performance assessment parameters can potentially provide a more detailed and comprehensive assessment.…”

Section: Sensitivity Analysis Of the Structure Of Annmentioning

confidence: 99%

A transfer learning based artificial neural network in geometrical design of textured surfaces for tribological applications

Mousavirad

Rahmani

Dolatabadi

2023

Surf. Topogr.: Metrol. Prop.

View full text Add to dashboard Cite

This study aims at introducing the potential to utilise transfer learning methods in the training of artificial neural networks for tribological applications. Artificially enhanced surfaces through surface texturing, as an example, are investigated under hydrodynamic regime of lubrication. The performance of these surface features is assessed in terms of load carrying capacity and friction. A large performance dataset including bearing load carrying capacity and friction is initially obtained for a specific category of textures with rectangular cross-sectional profile through analytical methods. The produced bearing performance are used to train a neural network. This neural network was then trained further by a minimal set of performance measure data from an intended category of textures with triangular cross-sectional profiles. It is shown that the resulting neural network performs with acceptable level of confidence for those intended texture profiles when trained with such relatively low number of performance data points. The results indicate that fast analytical methods can potentially produce a large volume of training datasets, which effectively allows for use of relatively lower number of training data sets from the intended category, where creating data for trainings can be more complex or time consuming. Use of transfer learning method in tribological applications and use of bearing performance parameters, as opposed to bearing design parameters, for training the neural networks are the major novel contributions of this study, which has not hitherto been reported elsewhere.

show abstract

Section: Sensitivity Analysis Of the Structure Of Annmentioning

confidence: 99%

A transfer learning based artificial neural network in geometrical design of textured surfaces for tribological applications

Mousavirad

Rahmani

Dolatabadi

2023

Surf. Topogr.: Metrol. Prop.

View full text Add to dashboard Cite

show abstract

“…ANN is the computational system that can be used to solve the critical problem occurring in the field of process identification, error filtering and product and design development [16]. ANN is used to model non linear problem and predict the responses for selected parameters from their training values [17,18]. Eventhough, it has some limitations (i) no specific rule for determining the structure of artificial neural networks, (ii) difficulty showing the problem to the network, (iii) ANNs can work with numerical information and (iv) problems have to be translated into numerical values before being introduced to the ANN.…”

Section: Artificial Neural Networkmentioning

confidence: 99%

Mathematical and artificial neural network model in composite electrode assisted electrical discharge coating

Elaiyarasan¹,

Satheeshkumar²,

Srikanth³

et al. 2022

Surf. Topogr.: Metrol. Prop.

View full text Add to dashboard Cite

Surface coating is covering the surface of workpiece with desired materials in order to improve the surface properties. Electrical discharge coating (EDC) is an electro thermal process, used for creating hard coating over the workpiece. EDM parameters play important role to improve the surface properties. In this present study, magnesium alloy is deposited using WC-Cu composite electrode by EDC. RSM is used to develop design matrix for carrying out EDC experiments. During experiments, compaction load, discharge current and pulse on time are controlled, whereas material deposition rate (MDR) and surface roughness (SR) are measured as response. The objective of this investigation is to predict the MDR and SR using neural network technique. ANN (Artificial neural network) model developed by back propagation algorithm is proposed in this study for predicting responses. ANOVA is conducted to identify the dominating parameter, which significantly affects the responses. Correlation coefficient between the ANN and RSM is 0.99, which is close to the unity for ANN. It was revealed that the prediction of proposed ANN was found to be excellent to the RSM model. MDR increased with increasing of discharge current and on time. SR decreased with increasing of applied load of the compaction, whereas when increase the current and pulse on time, SR increased.

show abstract

“…Step 2: The EF , SR , and MRR models are developed regarding process parameters by means of the ANN. 22,23…”

Section: Optimization Approachmentioning

confidence: 99%

“…Step 2: The EF, SR, and MRR models are developed regarding process parameters by means of the ANN. 22,23 The ANN approach is named as an advanced processing and modeling technique for the obtained data, in which the biological behaviors of the human brain are simulated and reproduced. The ANN requires a number of layers and neurons, which are connected to produce the neural network.…”

Section: Optimization Approachmentioning

confidence: 99%

An energy efficiency-based optimization of the rough honing process of hardened 5150 alloy

Van

Nguyen

2022

Proceedings of the Institution of Mechanical Engineers, Part E:

View full text Add to dashboard Cite

The honing operation is maned as an effective and efficient finishing solution to achieve better surface properties and machining efficiency. However, quality and productivity indicators are primary considerations of published investigations. In this study, the rough honing operation of the hardened 5150 alloy has been addressed and optimized to decrease the energy efficiency ( EF), surface roughness ( SR), and material removal rate ( MRR). The optimizing factors include tangential speed ( T), linear speed ( L), honing pressure ( P), and honing depth ( D). The artificial neural network (ANN) method is utilized to present the non-linear relations between the process parameters and performance measures. The Taguchi method is applied to select the best architecture of ANN models. The Entropy and the vibration and communication particle swarm optimization (VCPSO) algorithm is applied to compute the weight factors of response and find optimal data. The total honing cost ( THC) model is proposed to compute the honing expenses. The results indicated that the optimal values of the T, L, P, and D are 32.0 m/min, 8.0 m/min, 3.3 MPa, and 0.01 mm, respectively, while the improvements in the EF, SR, MRR, and THC were 15.17%, 44.7%, 134.19%, and 25.0%, as compared to the initial values. The optimizing technique comprising the ANN, Entropy, and VCPSO could be named as an effective and efficient method to solve complicated optimization issues. The optimal data could help the machine operator to improve the honing performances with regard to saving trial costs and essential efforts.

show abstract

Artificial neural network-based modelling of optimized experimental study of xylanase production by Penicillium citrinum xym2

Cited by 8 publications

References 27 publications

A transfer learning based artificial neural network in geometrical design of textured surfaces for tribological applications

A transfer learning based artificial neural network in geometrical design of textured surfaces for tribological applications

Mathematical and artificial neural network model in composite electrode assisted electrical discharge coating

An energy efficiency-based optimization of the rough honing process of hardened 5150 alloy

Contact Info

Product

Resources

About