A METHODOLOGY FOR THE OPTIMAL DESIGN AND CAPACITY EXPANSION PLANNING OF NATURAL GAS TRANSMISSION NETWORKS.

摘要

The multiperiod, multicomponent, nested design and capacity expansion problems in natural gas transmission networks are investigated. A general mathematical (mixed-integer, nonlinear) model is developed. The general nature of the model enables specialization to be made to adequately represent either the static design case or the dynamic case that involves capacity expansion. The two major components of a natural gas transmission system are pipes and compressors, and these components have both market and technological interdependencies.;In the second level, a decomposition technique is developed that enables the capacities of the interdependent components (pipes and compressors) to be expanded independent of the technological factors that make them interact. This level establishes certain conditions that justify the use of this single-component decomposition to the capacity expansion problem. The concept of variable starting regeneration points is introduced. This concept enables one to determine the optimal sequence of expansion of a compressor component if the construction date is delayed until a date later than the start of the planning horizon.;The third level uses the modifications of the first level, and the optimal single-component capacity expansion schedules of the second level to find the optimal locations and operating pressures of the compressors, and also the optimal capacity additions to the pipes only in terms of cumulative equivalent diameters.;The fourth level develops a search procedure that determines, using information from levels one through three, the actual designs for the main and loop pipes, and the length of loop additions during the expansion intervals.;The model is solved using a four-level hierarchical solution technique. The first level specializes the model to that of the static design problem. Simplifications and other modifications effected at this level serve two main purposes. First, they reduce the problem to one with a nonlinear objective function and linear constraints. Second, they are used in the lower levels of the solution process.;The methodology developed here is applied to a real (proposed) natural gas transmission system.