Prediction of Water Production Rate Using Fuzzy Logic Approach

Abstract

Abstract—this paper describes a fuzzy expert system approach for prediction of water production rate in a desalination plant. The goal of the study is to predict water production rate based on several inlet variables determined by steam flow, steam pressure, seawater temperature, and seawater flow. The data used were collected from plant history records of the operation department log sheets. These data were analyzed and a model was constructed using fuzzy expert system. In the proposed model, both input and output variables are parameterized and classified into several fuzzy sets. A set of fuzzy rules are constructed based on the knowledge extracted from the data collected. Once the rules are evaluated, the variables are defuzzified and converted into corresponding output variable (water production rate). The results showed that inlet variables of the plant could predict water production rate with 95% prediction accuracy.