Optimization of mobile oil and gas produced water treatment unit deployment logistics to achieve economic feasibility

Abstract

• Techno-economic assessment of on-site produced water desalination . • The concept of capacity factor is applied to desalination where demand is variable. • Systems level optimization of mobile desalination module deployment. • Multi-objective optimization considering economics and water reclamation . Development of unconventional oil and gas wells has resulted in large volumes of wastewater that require careful handling to minimize environmental risks. Common practice is to truck the wastewater from well sites to Environmental Protection Agency Class II underground injection control wells. Recently, on-site wastewater treatment followed by surface water discharge for downstream reuse has emerged as being viable. This work applies the concept of capacity factor to on-site oil and gas wastewater treatment. By accounting for intermittent wastewater treatment demands at individual well sites, the attributes of optimal on-site packaged treatment unit deployment logistics for oil and gas wastewater management are determined. The attributes considered are packaged unit capacity, deployment length, and the number of units deployed. This work explores different deployment logistics in Weld County, Colorado, United States to determine a set of Pareto optimal logistics from a techno-economic and environmental perspective. Ultimately, there exists a Pareto frontier because the marginal cost of water treatment increases as reuse increases beyond a critical threshold, in Weld County where the cost of business as usual operation (injection) is constant at $7.68 per m 3 produced. Generally, optimal deployment logistics, when accounting for capacity factor, utilize packaged wastewater treatment units sized at 100 m 3 per day capacity with deployment location reevaluated monthly. This article focuses on the importance of capacity factors at on-site treatment units and presents limitations of the study and the opportunity for the application of the presented methods to other case studies through utilization of the open-source model produced in this work.