Application of Artificial Intelligence-Based Methods in Estimating Water Consumption Productivity (Case Study: Sari City)

Today, the optimal use of water in various sections to increase productivity and reduce losses is one of the basic assumptions in consumption management. Therefore, in the present research, the concept of productivity in different water consumption sections, including groundwater, drinking water, and special attention to chemical properties in the sewage treatment plant in Sari City has been investigated. In each section, using the appropriate algorithm, the available data has been evaluated to measure productivity. In the sewage treatment section, the analysis of effective parameters in Sari City sewage treatment was performed in two methods Artificial Neural Network and Wavelet. The output results of these models showed that due to the high value of the R2 statistic, there is an acceptable and direct relationship between the measured and estimated characteristics. In the groundwater sector, according to the results obtained, the wavelet network performed better in estimating the desired variables than the ANN method. In the water distribution network section, the results of wavelet analysis and the outputs of the WaterGems software revealed that the deterioration of the studied water distribution network plays a significant role in losses and reduced productivity, in such a way that about 47 percent of the water entering the network is out of reach and wasted in different ways.