Guest Editors
Dr. Lai Chin Wei, University of Malaya, Malaysia.
Dr. Sanjay Mavinkere Rangappa, King Mongkut's University of Technology North Bangkok, Thailand.
Dr. Seyyed Mojtaba Mousavi, National Taiwan University of Science and Technology, China (Taiwan).
Summary
Response surface methodology (RSM) has been widely utilised by various professionals in analysis and optimises a broad range of applications. The major limitation of RSM is the incompetence embodied by non-controllable decisive factors. An artificial neural network (ANN) is a soft computing methodology which processes through the modification of network weights to achieve the desired output. However, there is no necessitate of any meticulous knowledge of physio-chemical parameters influencing the process. As a result, the ANN can develop a compatible non-parametric simulative model. From numerous studies, it was pointed out that ANN can develop significant models with minimal mean squared error (MSE) and high correlation values. The combination of ANN with RSM could overwhelm the drawbacks of response surface methodology in the non-adjustable variables in modelling. This technique relatively utilises less number of trials and can predict more realistic solutions. Thus, modelling using the combination of response surface methodology with artificial neural network could be able to explore and understand empirical correlation among the independent variables and response factors. This special issue is focused on the experimental design by central composite design with the aid of response surface methodology to study the optimisation of kinetic adsorption efficacy using several promising catalysts in various applications.
Keywords
Response surface methodology (RSM), Artificial neural network (ANN), Mean squared error (MSE), kinetic adsorption efficacy, Catalysts
Login
Register
Submit a Paper
Propose a Special lssue