The energy system is comprised of (but not limited to) electricity, natural gas, liq-uid fuels, nuclear, biomass, hydroelectric, wind, solar, and geothermal resources. Modeling of national freight and passenger transportation focuses on state-to-state travel; we consider both infrastructures (rail, highways, locks/dams, roads, ports, airports) and fleets (trains, barges, trucks, personal vehicles, airplanes, etc.), and there may be different kinds of fleets for each mode (e.g., diesel trains and electric trains or conventional and plug-in hybrid electric).
The figure above captures the scope of our modeling effort. The transportation and energy systems interact mainly at two different stages: operation and investment. At the operational level each system needs to satisfy its demand with the existing capacity. However, operation of the two systems, and ultimately investment, are interdependent; while the transportation sector demands energy in the form of fuel, the energy sector requires the movement of raw bulk energy sources (e.g. coal or natural gas for thermal power plants). At the same time, the cost of meeting those reciprocal demands has an impact on final prices for energy and transportation. The ever-growing public need for energy and transportation creates the necessity to invest in new capacity. Given the potential for increased coupling between energy and transportation, it is apparent that better designs of both can be achieved if these designs are performed together.
One of the core features of this approach is its multiobjective nature. The final objective is to find a Pareto front of non-dominated solutions. Rather than a predetermined hierarchy or weighted system before the optimization takes place, the trade-offs can be analyzed a posteriori, giving decision makers and general public a solid background to determine where their efforts should be focused on.
The objectives are grouped in three distinctive categories: cost, sustainabilityand resiliency.
A multiobjective evolutionary algorithm is proposed as an approach to efficiently solve the problem described above and produce an approximation to the Pareto front. It has been conceived in a modular fashion, to allow the parallel development of each one of its components. At the highest level two levels can be distinguished, as depicted in the following figure.
The high-level optimization is commanded by the popular NSGA-II algorithm. Each individual within a given population is characterized by a minimum level of investment that is to be enforced.
These minimum investments are then passed on to the low-level optimization, which captures the modeling of the energy and transportation systems formulated as a linear program. Fast decomposition methods are applied to enhance the solution time of this optimization and produce the investment portfolio for that individual solutions as well as the flows representing how the system is operated.
Once the portfolio is determined, metrics for sustainability and resiliency are calculated. Those metrics can range from simple linear functions of investments and flows to multiple contingency-based scenarios.