(361b) Towards an Integrated Field Wide Production Optimization Approach

Integrated Asset Modelling (also known as Integrated Production System Modelling ‐ IPSM) is the modelling and simulation of an entire production facility consisting of both subsurface and surface elements. The intent is to holistically capture the complex interactions between the individual components of the system.

Historically asset modelling has entailed discrete reservoir, wells and facility models; however this mindset is slowly altering towards integrated asset modelling for production forecasts and optimisation whereby both system imposed constraints i.e. fluid flow from the reservoir and surface handling constraints can be considered simultaneously. This systems based approach is critical and reflects the current industry mind set. The full benefits of the integrated simulation tools employed within industry are not fully realised due to a number of
limitations; complexity of the reservoir models and hence limited usability in day to day production optimisation, overcoming the 'silo' discipline mentality that is common within interdisciplinary surveillance teams, simplification of integrated models is often in tension with accuracy and lastly, the limited capability of current tools to apply multi objective optimization analysis.

With these challenges, it can often be difficult to justify the effort and investment required to develop and deploy an integrated asset model. Despite these practical challenges, integrated asset modelling, if executed correctly can provide one of the best means of optimising production throughout field life with direct realized benefits such as extended field life and sustained production uptime.

The main objective of this research study is to understand the opportunities and limitations with the current existing industry integrated packages and develop a viable, representative alternative that is applicable for day to day surveillance and medium to long term field production optimisation. This research also aims to formally outline the drivers for integrated asset modelling followed by a review of the most common integrated asset modelling tools on the market in addition to review of the optimisation and reservoir modelling simplification methods available in literature. An integrated asset field model was developed including secondary recovery water injection as a test case.

Proxy models have been developed using proprietary HDMR meta modelling software, polynomial development and machine learning techniques at Imperial College London which was extended to the use of full physics reservoir models and associated sensitivity analysis.

These surrogate models have also been used in further heuristic-based optimization frameworks (including Genetic Algorithm optimisation of developed Artificial Neural Network surrogates) and sensitivity analysis with very positive results in comparison to the field wide simulators thereby presenting a viable alternative offering significant advantages in simulation time.