A novel image processing technique to evaluate biodiesel wastewater for recovery, recycle and reuse towards zero liquid discharge approach

Abstract

Abstract Water is an essential resource which requires to be preserved for the living beings. Wastewater generated from both domestic and industrial applications needs to be recycled in order to meet the increasing demand in everyday life. The present study focuses on the quality assessment of biodiesel wash water which evolved in biodiesel production that demands a technology to recover, recycle and reuse. The various parameters that evaluate the quality of wastewater before and after treatment are pH, turbidity, COD, BOD, DO etc., While several methods of detecting the water quality parameters are available a novel, reliable, quicker and dry laboratory technique for evaluating pH and turbidity using image processing is developed. Wet lab pH and turbidity data of wash water obtained by washing rice bran oil biodiesel, groundnut oil biodiesel and palm oil biodiesel has been used as a basis for developing the module. Images of the various wash water samples were also captured. Image processing classification is performed using convolution neural network. A Graphical User Interface is developed for pH and turbidity monitoring of the given biodiesel wash water image. The pH and Turbidity parameters were compared with conventional test results and the average deviation was estimated. Validation of the proposed module for its precision is done using chi-square test.