Optimization of Injection Scheduling in Geothermal Fields.

摘要

This study discusses the application of algorithms developed in Operations Research to the optimization of brine reinjection in geothermal fields. The injection optimization problem is broken into two sub-problems: (1) choosing a configuration of injectors from an existing set of wells, and (2) allocating a total specified injection rate among chosen injectors. The allocation problem is solved first. The reservoir is idealized as a network of channels or arcs directly connecting each pair of wells in the field. Each arc in the network is considered to have some potential for thermal breakthrough. This potential is quantified by an arc-specific break-through index, b/sub ij/, based on user-specified parameters from tracer tests, field geometry, and operating considerations. The sum of b/sub ij/-values for all arcs is defined as the fieldwide breakthrough index, B. Injection is optimized by choosing injection wells and rates so as to minimize B subject to constraints on the number of injectors and the total amount of fluid to be produced and reinjected. The study presents four computer programs which employ linear or quadratic programming to solve the allocation problem. In addition, a program is presented which solves the injector configuration problem by a combination of enumeration and quadratic programming. The use of the various programs is demonstrated with reference both to hypothetical data and an actual data set from the Wairakei Geothermal Field in New Zealand. (ERA citation 13:042967)