Improved Multiple Linear Regression for Forecasting Water Quality of Vegetable and Fruit Wastewater

The waste and wash waters from vegetable and fruits process industry are categorized with solids and organic component which needs treatment for meeting regular standards in use. Every treatment is assigned with minimizing postponed solids, whole phosphorus, biochemical, chemical oxygen and nitrogen demanded. From produced data, that is probable for developing prediction model to every tested water treatment. The methods for predicting post treatment water quality and created by Improved Multiple Linear Regression (IMLR), that is enhanced through Generalized Structure of Group Method in data handling methods (GSGMDH). The selection of IMLR and generalized architecture of group technique of data handling methods because of capability of methods for generating equations to utilize. The huge variability and difficult nature of waste water quality parameters are difficult in representing linear methods, but it is perfectly matched for group model of data handling method. The method provided an significant tool for end user in choosing correct treatment depended on actual characteristics of waste water and needed standards for treating waste.