Prediction of Daily Water Demand of the Shiraz City Using Neural Network and Honey Bee Colony Optimization Algorithm

Access to safe drinking water is one of the most important human needs and citizenship rights. For this reason, the supply, transmission, treatment, and distribution of sanitary drinking water to meet the water needs of urban and rural subscribers is one of the primary tasks and priorities of any government. Predicting water demand in water supply and distribution systems will be of great help to managers related to water supply, to manage and prevent crises and water supply planning, service and maintenance of equipment and facilities, culture, information and so on. In this study, a combined method based on neural network methods and bee colony optimization to predict drinking water demand and health of Shiraz is presented. The purpose of this study was to improve the accuracy of water demand forecasting using the neural network method. Parameters considered for modeling water demand forecasts include past information on water demand, air temperature, population, wind, and date. The data used to train the neural network included 10 years from 1988 to 1997. In order to verify and evaluate the performance of the proposed method, water demand in 1998 and April 1999 has been predicted and compared with real statistics. Based on the obtained results, it was concluded that the proposed method was able to adequately predict water demand. The proposed method has good accuracy and the deviation of the water demand forecast in the worst conditions has reached one percent, which is an acceptable amount. Statistically, the results obtained using the MAPE parameter were compared with previous studies, and from this perspective, the proposed method is reliable and has good efficiency in predicting the water demand of the Shiraz city system.