(302e) Valuation of Flexibility in Energy Conversion Networks | AIChE

(302e) Valuation of Flexibility in Energy Conversion Networks

Authors 

Mousaw, P. - Presenter, University of Notre Dame
Kantor, J. - Presenter, University of Notre Dame


Uncertainty in raw material costs, demand, and sale price of finished goods makes a process industry's goal of maximizing profit a challenging optimization problem. With recent fluctuations in energy fuel prices, this is especially true of energy conversion networks. These networks must meet demand while having to satisfy many constraints, such as total power output equipment can produce, types of fuels that can be use, environmental regulations, as well as others.

This paper develops methodology for valuation of flexibility in energy conversion networks. The methodology is intended for use in campus scale utilities with complex energy requirements, fuel sources, and operational flexibility. Historical fuel price data is collected for several fuels of interest. Several stochastic process models are used to fit this historical fuel price data. These fits are used to generate fuel cost uncertainty models.

A class of bilinear models suitable for the optimization and risk management of commodity energy conversion networks introduced in an earlier paper is used to model several energy conversion networks. This class of bilinear models uses first and second law principles from finite-time thermodynamics to analyze energy conversion networks. These bilinear models coupled with a set of fuel price realizations using the cost uncertainty models to calculate the deterministic economically optimal process operation conditions.

Using a Monte Carlo method, economic optimization is performed over many fuel price realization sets. These results are aggregated to obtain an expected value of the option of fuel choices in the energy conversion network. Several examples of varying complexity are presented.

In summary, this paper presents 3 main results:

? Uncertain fuel price models are constructed using historical data and stochastic process models.

? Economic optimal process operations is determined using these uncertain fuel price models.

? Determination of the value of fuel and operation flexibility using Monte Carlo programming techniques.