Abstract
Production optimization for offshore oil and gas production is in general a challenging task, even at fields operated "Smart" with continues 24/7 optimization, due to the intrinsic complexity of the domain. In this paper, we present an intelligent multi-objective-control-software approach for the next generation of Smart Fields.
At the DONG Energy E&P operated Siri area, which we use as a test case, several optimization studies have shown that an increase in production throughput is possible, if the comfort zone, i.e. the band between the actual and the maximally possible production level, is reduced dynamically. Maximal production will be obtained when the comfort zone meets the minimally required margin that ensures a safe and stable production in all constraining systems of the installation. This can be hard for the control operators to achieve with the present complex dynamic production configurations.
In our approach, we focus on intelligent online control to minimize the comfort zone by pushing the production towards the process constraints which always have to be satisfied. The new intelligent online multi-objective control system is implemented as a stratified multi-agent system allowing control concerns to be dynamically introduced, changed or removed without the need to modify or inspect the existing control system. The stratified approach supports multi-objective optimization in all layers, i.e. in contexts of strategy, tactics and operation. Optimization conflicts are dynamically identified and propagated to a higher layer. The "irony of automation" predicts that more advanced automation systems require more tacit knowledge; an essential property of any advanced control system is, therefore, the capability to identify and explain optimization conflicts well in advance.
In this paper, we demonstrate that it is possible to continuously minimize the comfort zone and thereby gain higher production throughput by using intelligent online multi-objective control, even at fields with complex configurations.
