| dc.description.abstract | Satisfaction of global energy and water demands in a sustainable manner is a great challenge, which is often amplified by the high interdependence between energy and water networks. This interdependence is captured in a Water-Energy Nexus (WEN) and can result in redundant or excess resource withdrawal. Sustainability, survivability and resilience are three key concepts in complex interconnected networks such as WENs. These need to be analyzed and quantified for economic WEN design. WEN interdependence has high impact during disruption events, impairing the network operability and profitability. For example, internal connectivity disruptions in a power network affect the performance of a water distribution network, and vice versa. In addition, prolonged demand and generation disruptions can put the long-term survivability of an interconnected supply chain at risk. Recent natural disasters (2021 Texas Freeze, Hurricane Harvey) and COVID-19 pandemic-induced lockdowns revealed such vulnerabilities. In this dissertation, a framework is presented to (i) ensure minimal redundancies and cost-effective regional WEN design, (ii) guarantee network resilience against connectivity disruptions with minimum additional costs, and (iii) ensure economic survivability during demand disruptions.
To address sustainability, a graph-theoretic approach is proposed defining a nexus as a directed bipartite graph with water and energy flows. The network representation allows the decomposition of a complex nexus into its essential and redundant components based on the intensity of the generating technologies. It is shown that for specified external grid demands, the optimal nexus configuration with minimum water and energy generation is the one without any redundant subnetworks. A novel WEN diagram is introduced to represent networks and a graphical pinch method is developed to identify and eliminate redundant subnetworks. This leads to minimum generation/resource utilization, while also taking into account for matching restrictions and water quality specifications. The method can be used as a screening and targeting tool for optimal technology combinations with minimum redundancies. Furthermore, a WEN superstructure optimization-based approach is developed to find optimal WEN infrastructures while considering wastewater reclaim and reuse, multiple resource types, varying water quality, and facility location-allocation, via a mixed-integer nonlinear programming (MINLP) model.
To address resilience, the operational and economic performances of a WEN during connectivity disruptions are analyzed. To this end, minimum cost of resilience (MCOR) and operation-based resilience metrics are introduced and utilized to identify critical connections in interconnected networks. MCOR corresponds to the minimum additional infrastructure investment that is required to achieve a certain degree of resilience. To guarantee MCOR for grass-root or retrofitting applications, the metrics are incorporated in a multi-scenario mixed-integer linear program (MILP) that accounts for resilience in the design phase of interconnected networks. Increasing immunity to connectivity disruptions leads to increased investments allocated in excess system capacities or higher dependence on external supplies.
To address survivability and predict the economic performance of supply chains against demand disruptions, the concept of economic survivability (ES) is introduced and incorporated in a mixed-integer nonlinear program (MINLP). ES is the ability to maintain a net positive economic worth, or at least keeping it above a certain threshold, in the presence of sudden and prolonged disruptions that drastically reduce the product demands, prices, resource availability or others. It is observed that, maximizing ES leads to systems with higher return-on-investment (ROI) and profitability. However, for multi-regional, distributed and interdependent supply chains, a more balanced distribution of investment portfolio is important to improve the local survivability of each region, but it comes at the expense of overall profitability. The effect of overdesigning for the event of increased demands is also explored. Higher demands satisfied lead to lower economic survivability under demand decreases, so the decision-makers should balance the trade-offs between survivability and excess demand satisfaction by thoroughly assessing the probability of positive and negative demand fluctuations. | en |
