Stochastic, Multiobjective, Mixed-Integer Optimization Model for Wastewater-Derived Energy

Abstract

Assessing investment decisions in wastewater treatment can be difficult given that operators face large fixed and variable costs as well as large amounts of uncertainty in the amount of wastewater inflow as well as other factors. Decision makers can solve these problems by implementing operations research models that capture environmental engineering and economics aspects of the problem. Perfect decisions require perfect information but decisions makers can use stochastic models to hedge their decisions against an uncertain future. This paper presents a stochastic, multiobjective, mixed-integer optimization model for a wastewater treatment plant (WWTP). The WWTP can convert the solid end product from wastewater into methane to produce revenue and renewable energy credits. Alternatively, the solids can be used as biosolids for land application, agricultural markets, or compressed natural gas transportation markets. This stochastic optimization model explicitly considers probabilistic information, and the expected value of perfect information and the value of the stochastic solution provide important information for decision makers. As such, the model considers many aspects of the Smart Grid such as integration between energy and transportation, electricity generation by atypical prosumers (producer and consumer), and overall system planning to reduce negative environmental externalities, to name a few. It is shown that a significant trade-off exists between operational and investment costs and the associated carbon dioxide emissions. The WWTP could reduce the amount of carbon dioxide emissions but the operational and investment costs would be increased. One of the results determined in this paper is that to reduce 1 t of carbon dioxide equivalent emissions (given average energy consumption levels) requires $36, $173, or $371 per day when the range of carbon dioxide equivalent emissions is (177,202] t