Practical real-time optimization for energy efficient water distribution systems operation

By: Salomons E., Housh M.
Published in: Journal of Cleaner Production
SDGs : SDG 07  |  Units: Social Sciences  | Time: 2020 |  Link
Description: The production, treatment and delivery systems of drinking water and wastewater is one of the largest energy consumers i n the US with about 4% of the nation’s power consumption. Roughly 80% of the water treatment and distribution costs are associated with electricity, mainly for pumping. Increasing the efficiency of drinking water pumping systems could benefit both the energy- and water-sectors. Despite the advancement of optimal pump scheduling technology, most water utilities are relatively small, and thus lack the funds, hardware and technical personal to support the use of sophisticated and computer intensive pump optimization programs. This study presents a simple and practical model predictive control methodology for real-time pump scheduling. This methodology can be deployed on a standard hardware (e.g., PLCs in pumping stations), which is currently in use by most water utilities. As such, it provides optimal pump scheduling benefits without necessitating large investment in new computational hardware (e.g., advanced controllers). The proposed methodology reduces both the energy consumption (by selecting the most efficient pumps’ combinations) and the operation cost (by optimizing the pumps’ operation according to electricity tariff periods). The results show that our practical methodology, which could be implemented in simple controllers, can provide near optimal decisions comparable with sophisticated optimization methods that require advanced hardware. To the best of our knowledge, there is no available methodology with such capabilities which is specifically designed for local control schemes. Thus, the novelty of this study is the utilization of this optimization methodology on a simple PLC hardware. Our results are of high importance for both academic and practical reasons, as it shows that the proposed methodology could be a kernel for a low-cost pumps’ optimization technology. © 2020 Elsevier Ltd