Optimisation and Modeling Approaches for the Textile Industry Water Treatment Plants

Abstract

The textile industry is considered a major pollutant source among all the industrial units because this industry intensively uses a variety of chemicals for the pre-treatment and processing of fibres, wool etc. The wastewater generated from textile processing plants has a complex chemical composition and abnormally elevated physical characteristics. Various chemicals such as organic dyes, bleaching agents, fixing agents etc., are used to upgrade the characteristics of the finished textile materials. A number of methods such as adsorption, coagulation, electro-Fenton oxidation, membrane separation and biological degradation are followed to eliminate the undesirable components from the outlet stream. The success of the treatment process depends on understanding the underlying mechanism of mass transfer by diffusion, the kinetics of pollutant removal and hydrodynamics of mixing. Modeling represents a process by mathematical equations, which comprises the variables affecting the process performance. The model equation shows the relationship between the input and response variables in a process. Model of a process helps to simulate the conditions and understand the robust behavior of systems. The optimisation is a mathematical approach to identify the best condition for a process. The objective of optimisation is to minimise the operating cost or maximise process efficiency. In the wastewater treatment domain, modelling and optimisation are helpful to understand the pollutant removal rate, demarcate the major variables affecting the process efficiency and identify the range of operating conditions. Textile wastes have high salinity, prohibitive total dissolved solids and residual organic dye compounds. The important models developed for a treatment plant are the mass transfer, kinetic, adsorption, and process models. The mass transfer model gives an insight into the rate of diffusion of pollutants in an aqueous medium. A kinetic model explains the rate at which undesirable compounds are removed from wastewater and elucidates the effects of temperature on the process. The process models are used to realise the important variables affecting the process efficacy. The process model explains the interactive effect and linear effect of variables on the response variable. The modelling tools used in textile treatment plants are response surface method (RSM) and Artificial Neural Networking (ANN). RSM is the most widely used method to develop the model equations and optimise the process. The second order quadratic models developed by the RSM method interpret the effects of parameters on the response variables. RSM utilises the experimental design method to develop the complete model expression for the operation with the least number of experiments. Irrespective of treatment strategies for textile wastewater such as coagulation, adsorption, electrochemical oxidation, bio-degradation, ozonation, photolytic degradation and membrane filtration, the RSM is used as a versatile method for building the model and process analysis. The quality of the model equations is tested by statistical tools such as ANOVA table, fit statistics table, 2-D contour graph and 3-D response surface plot.