Design and development of artificial neural networks for depositing powders in coating treatment

Jean, Ming-Der; Liu, Chyuan-Du; Wang, Jen-Ting

doi:10.1016/j.apsusc.2004.10.041

Cited by 40 publications

(17 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Various authors used ANNs to model the unknown relationship between the process parameters and process responses, and to model unknown relations of various manufacturing engineering systems or behaviours, without imposing an assumption regarding the relationship form (e.g. Cook and Shannon 1992;Ko et al 1998Ko et al , 1999Galantucci et al 2000;Hinduja et al 2000;Monostori and Viharos 2001;Hsieh and Tong 2001;Hsu et al 2004;Galantucci et al 2004;Tong et al 2004a, b;Liau and Chen 2005;Jean et al 2005;Hsieh 2006;Dhara et al 2008;Noorossana et al 2008;Chou 2006, 2008;Salmasnia et al 2012), including modelling of a dynamic system (Su et al 2005;Su and Hsieh 1998). When ANNs are combined with Taguchi method, typically, the quality characteristics of a product (i.e.…”

Section: Multiresponse Optimisation Based On Artificial Neural Networkmentioning

confidence: 99%

Review of Multiresponse Process Optimisation Methods

Šibalija

Majstorovic

2015

Advanced Multiresponse Process Optimisation

View full text Add to dashboard Cite

The review of methods used for multiresponse process parameter design and similar multiobjective optimisation problems is discussed in this chapter, implying the following classification: (1) conventional methods based on statistical or mathematical techniques: (i) experimental design techniques (response surface methodology, Taguchi's robust parameter design and related approaches), and (ii) iterative mathematical search techniques; (2) non-conventional methods based on artificial intelligence techniques: (i) fuzzy logic, (ii) artificial neural networks, (iii) metaheuristic search techniques (genetic algorithm, simulated annealing, particle swarm optimisation, ant colony optimisation, tabu search, and recent evolutionary algorithms such as artificial bee colony algorithm, biogeography-based optimisation, and teaching-learning-based optimisation), and (iv) expert systems.

show abstract

Section: Multiresponse Optimisation Based On Artificial Neural Networkmentioning

confidence: 99%

Review of Multiresponse Process Optimisation Methods

Šibalija

Majstorovic

2015

Advanced Multiresponse Process Optimisation

View full text Add to dashboard Cite

show abstract

“…Therefore ANN has already been widely used in research on material processing and properties as an alternative to traditional prediction and control instruments based on statistical analyses and regression models. In the field of coating, there are also some successful applications [11][12][13]. Wu et al has utilized ANN to model the plating rate and phosphorus content in the coating of binary Ni-P plating [14].…”

Section: Introductionmentioning

confidence: 99%

Artificial neural networks implementation in Ni-Cu-P ternary coating: Investigation of the effects of bath stabilizers

Luan

Zou

2012

2012 8th International Conference on Natural Computation

View full text Add to dashboard Cite

artificial neural networks (ANN) were implemented to model a complex chemical reaction system: process of electroless plating of Ni-Cu-P alloys. This model was developed to simulate and predict plating rate as a function of amount of stabilizers added in the bath. The neural network was established with three layers and trained by the back propagation learning algorithm. The training and testing data were obtained by experiments. The simulation results of the neural network coincided well with the experimental value. Hence artificial neural network is a reliable method to optimize the process parameters of Ni-Cu-P coating.

show abstract

“…Back propagation network have hierarchical feed forward network architecture, and the outputs of each layer are sent directly to each neuron in the layer above. (Ming-Der Jean , Chyuan-Du Liu, Jen-Ting Wang, 2005) Back propagation network are trained by repeatedly presenting a series of input/output pattern sets to the network. The neural network gradually 'learns' the governing relationship in the data set by adjusting the weights between its neurons to minimize the error between the actual and predicted output patterns of the training set.…”

Section: Structure Of An Ann Model For Coating Thicknessmentioning

confidence: 99%

“…During the process, the polymer undergoes very complex thermo-mechanical transformations inducing strong changes in the physical properties of the material. (Ming-Der Jean, Chyuan-Du Liu, Jen-Ting Wang, 2005) A high-quality extrusion is essentially characterized by a precisely-regulated output volumetric flow; this can be achieved by finely regulating the temperature and the pressure of the die at the output of the extruder (Noriega, del pilar, & Rauwendaal, 2001). (Hulya Kacar Durmus, Erdogan Ozkaya,Cevdet Meric, 2004) Paste extrusion is widely used in the food, chemical, ceramic and pharmaceutical industries.…”

Section: Introductionmentioning

confidence: 99%

Prediction of the coating thickness of wire coating extrusion processes using artificial neural network (ANN)

Kozan¹,

Çirak²

2009

MAS

View full text Add to dashboard Cite

This paper presents a new method of modeling the nonlinear parameters of a coating systems base on neural Networks with artificial neural network neurons. Artificial neural networks (ANNs) are a new type of information processing system based on modeling the neural system of human brain. The wire coating thickness and quality depend on the wire speed, polymer viscosity, polymer melt temperature and the gap between the wire and exit end of the die. In this paper, results of experimental investigation are presented by comparing the coating quality on galvanized mild steel wire using EP 58 PVC molten is used as the coating material in a wire coating extrusıon unit at different extruder temperatures and extruder speeds.The coating thickness and quality are also discussed for different wire speeds of up to 15 m/s. A three layer back propogation artificial neutral network (ANN) model was used for the description of wire coating thickness.On comparing the experimental data, the predictions the ANN model predictions, it is found that the ANN model is capable of predicting the coating thickness. The neural network model shows how the significant parameters influencing thickness can be found. Inthis studies, a back propagation neural network model is developed to map the complex non-linear wire coating thickness between process conditions .

show abstract

Design and development of artificial neural networks for depositing powders in coating treatment

Cited by 40 publications

References 9 publications

Review of Multiresponse Process Optimisation Methods

Review of Multiresponse Process Optimisation Methods

Artificial neural networks implementation in Ni-Cu-P ternary coating: Investigation of the effects of bath stabilizers

Prediction of the coating thickness of wire coating extrusion processes using artificial neural network (ANN)

Contact Info

Product

Resources

About