| Popis: |
Gathering significant volumes of natural gas involves connecting to wells in different oil and/or gas fields that are usually spread across large geographical areas. Supplying processed gas and associated liquids, such as liquefied petroleum gas (LPG) and natural gas liquids (NGL) to the consumers often requires an extensive network of degassing stations, compressor stations, pipelines and gas processing plants, as well as storage and shipping facilities for the liquids. The operation of such natural gas networks involves many degrees of freedom impacting profitability, efficiency and environmental performance, which creates significant scope for optimisation. This work presents novel methodologies making use of detailed models for optimizing integrated production, processing and supply chain operations of natural gas networks. Despite the underlying complexity of the system, the application of rigorous mathematical optimisation to detailed models of an entire gas network is now computationally tractable and can ensure that the solution obtained meets all constraints. Using the vast natural gas network of Basrah Gas Company (a Shell Joint-venture in Iraq) as a case study, optimisation approaches to find optima to profitability, production and flexibility problems of natural gas networks have been developed. The magnitudes of the resulting improvements in optimised operations are significant. A Gradient Ranking method has been developed and tested for determining the significance of the decision variables to the objective function and thereby reducing problem complexity. Methods have also been developed to assess the relative economic impact of the different products (dry gas, propane, butane and natural gasoline) in relation to optimal profitability with varying market prices. Finally, a multiperiod optimisation (MPO) approach has also been developed to maximise the overall profit for multiple periods by varying, in each period, the process operating variables, sales rates and inventory levels. The results show that the MPO approach results in significantly improved profitability compared with that achieved via the sequential single-period optimisation approach. Open Access |
