Integrated dynamic drilling hazard identification

With this new knowledge we can contribute to safer well-placement and increased production from the reservoir through a better description of where the oil is produced from what happens with the rock during production. The main focus since the startup in April has been development of methods for forward modeling of the seismic wave field. Results from this work will be concluded in a paper on forward modeling and the Born approximation. This paper is aimed to be completed in Q1 2016. Furthermore, these tools will be used for the remainder of the project when studying induced seismicity. A PhD project report has been submitted and presented at University of Bergen, and the project has been approved by the committee in the petroleum geoscience group. The report includes a timeline for expected progress and important milestones. Leading up to the presentation at UiB, a workshop was held in Stavanger (Schlumberger Stavanger Research). This provided a valuable opportunity for all project partners to meet and exchange ideas. In the PHD project, the candidate has developed methods for modelling of seismic waves in both elastic and viscoelastic 3D heterogeneous media. The methods are especially efficient because waveforms at a certain point in time can be computed independently of all other times. This makes source inversion procedures such as moment tensor inversion very efficient as we are mainly concerned with transmitted waves around the direct arrival. Two papers with the PHD candidate as main author will be published on the modelling methods. Furthermore, we have applied the modelling and inversion methodology to a perforation shot at the study field. This is now used as a reference for a sensitivity study where we investigate the limitations of the inversion. Work during 2017 in this project has been primarily focused on two topics. The first topic is computation of ray-Born integrals for efficient forward modelling and the second part is the implementation and testing of inversion methods for microseismic event location and source characterization in 3D velocity models. Three journal paper and one conference paper is planned for the forward modeling studies. Two papers have been published in Geophysics and the work with the third paper is in progress. The novel event location method has been tested on field data and compared to the current detection methods used by the operator. The results will be presented to the operator, while the methods used will be IP-protected by Schlumberger. In 2020 the work i the project has centered around finishing the work and publication of the third paper in the thesis on 3D visocelastic waveform inversion and compilation of the other parts of the project into a final report.

Prosjektet har tilført prosjektpartnerne ny kompetanse om dynamiske forkastnings- og sprekk-endringer observert på repererte seismikkmålinger. Det ventes videre at denne kompetansen brukes for videre utvikling av produkter og tjenester innen reservoarkaraktrisering og reservoarmonitorering.

Well integrity for the total lifespan of an oil field is important for the environment and for the economics of an oilfield. When a field is produced the mechanical state in the reservoir and the overburden changes with time. These changes are important for the integrity of existing wells as well as for the planning of new wells. A Sintef study found that 20-30 % of all wells have well integrity failures. We want to combine active and passive seismic for monitoring to provide a 4D characterization of the overburden, and potentially the reservoir. Together they will give a better understanding of the mechanical changes in the overburden and the reservoir and will allow us to derive drilling parameters that better represent the stress state in subsurface rock. The large compaction and overburden subsidence at Ekofisk is an ideal test case for the innovation. In addition, the PRM system at Ekofisk can be run in a passive mode detecting seismic events in the subsurface. The PRM also ensures high repeatability of the time-lapse seismic surveys. This will provide a field size laboratory for surface passive seismic since the dataset can be partitioned into several different geophone array configurations, less than a full PRM system, which can be studied. In this proposal we investigate how to improve reservoir sweep efficiency and drilling hazard predictions by means of active and passive seismic monitoring. These seismic data types will be applied to characterize stress state changes and mechanical property changes in the subsurface which will provide improved input for drilling. The innovative step is to tightly integrate epicentre events from the passive seismic network with novel time-lapse attributes to dynamically update the coupled geomechanical/reservoir flow model. Thereby compaction effects in the reservoir and potential reactivation of fault and fracture systems may be detected by using the type of DNA search method developed in phase 1 of the project.