Estimate the Impurity and Purify Water by ANN (KSOM)

Water supply entities have the responsibility to supply clean & safe water as required by the consumer. It’s therefore necessary to implement mechanism & system that can be employed to predict both proposed short term & long term water demand.Water treatment includes many complex phenomena, such as coagulation & flocculation. These reactions are hard or even impossible to control satisfyingly by conventional methods. The increasingly growing field of computational intelligence techniques has been proposed as an efficient tool in the modelling of dynamic phenomena. Traditionally, jar tests and operators' own experience are used to determine the optimum coagulant dosage. However, jar tests are time-consuming and less adaptive to changes in raw water quality in real time. When an unusual condition occurs, such as a heavy rain,the storm water brings high turbidity to water source, and the treated effluent quality may be inferior to water quality standards, because the conventional operation method can be hardly in time to adjust to the proper dosage. An optimal modelling can be used to overcome these limitations. That is why we have tried to apply the intelligent methods to control wastewater purification. In this paper, we describe intelligent technique that is artificial neural network (ANN) KSOM model to correlate the real data to the SOM (self organizing map) in which Variable selection (based on correlation analysis) and dimension reduction (based on principal component analysis) are used for data pre-processing. The significance of the proposed neural network-based approach is that it can model the unknown non-linear relationship between Coagulation & sedimentation measurements without requiring any prior knowledge of their inherent relationship. The goal primarily is to develop a reliable neural network based model, which can be extended for application and adaptive network-based system models.