Analyzing Waterflood Communication
Posted by D Nathan Meehan June 20, 2013

Analyzing Waterflood Communication: In this month’s blog, Baker Hughes reservoir engineering consultants Shelia Marsh and Atul Kshirsagar contribute to the blog. I also appreciate Neil Meldrum and Edwin Jong’s input.
Sweepscan improves waterflood performance

One key to optimizing oil recovery and determining production potential of waterflooded fields is through reservoir surveillance. Geological studies alone cannot quantify potential increases in reserves and oil rates achievable through optimizing existing waterfloods. Similarly, reservoir models based on these studies cannot a priori optimize injection or completion strategies. Dynamic reservoir simulation, incorporating geological knowledge, is usually used to understand fluid flow in the reservoir however; as a precursor to simulation it is strongly recommended that combined statistical and classic waterflood surveillance techniques are used to interpret reservoir fluid flow.

SweepSCAN™ is a tool developed in-house by Baker Hughes Incorporated in which communication analysis (CA) is performed to gain a preliminary understanding of communication between injectors and producers using only historic production and injection data. This process only takes a few days to carry out for even the most complex reservoirs. Understanding gained from the analysis can then be translated into 3D geological and reservoir engineering models and used by integrated subsurface teams to optimize waterflood economics.

Early detection of well communication is critical because it provides the first indication of well interactions. These, in turn provide insight into reservoir geology and heterogeneity. The resulting information can be used to identify target locations for further drilling, well conversions from production to injection or changes required to injection rates across the field to improve sweep.

Figure 1 Waterflood Response in a Producer

The process can also provide information regarding potential flood expectations, direction of fluid movement, reservoir storage capability and response time between injectors and producers. Furthermore, cycling of water or CO2 at the producers (perhaps due to high permeability layers) can be identified and reduced to optimise flood performance.
Figure 1 shows water injection and water production rates. The Energy Method employed uses algorithms to measure the effect of water injection upon producer response and to indicate the strength of this response between producer and injector well pairs.

This is applied to the whole field resulting in CA Rose diagrams (Figure 2) where strong communication is shown by the yellow and green markers. The summary diagram in the top left corner is a superposition of all the markers calculated; indicating that the principal direction of the communication is NE-SW.
Figure 2 Output from Communication Analysis

Bubble maps highlight the location of wells that have injected the most volume and show any association with high cumulative oil production or early water breakthrough. Strongly communicating well pairs are identified with CA. These are analysed against bubble maps and production data to validate possible communication channels. This is an important step because the wells showing high water production relative to oil production are most likely to be the strongest communicators responsible for the majority of water cycling in the field. It can also help identify field wide pattern or trends.

Examination of cumulative water produced and cumulative water injected (Figure 3) helps to identify groups of wells with similar trends (highlighted in red). In this example all three groups are aligned in the same orientation, corresponding to the NE-SW fracture trend in the field. Further analysis of the Group 1 wells identified here is performed to determine communication strength and to validate distinct flow paths.
Figure 3 Bubble Map of Cumulative Water Production and Injection

The correlation of the production and injection rates for the Group 1 wells depicted in Figure 3 is shown in Figure 4. A strong correlation is shown with arrows.

Figure 5 shows a case where there is no correlation between the injection rate and the production rate trends for a wells pair located NW of Group 1. This provides compelling evidence for a fracture system being present between these wells oriented NE-SW.
Figure 4 Injection and Production Data 1 (Group 1)

Once validated, the communication pathways for the whole field are mapped out using a color code and width scale to indicate communication strength as shown in Figure 6; this provides a visual map of the fluid flow in the field that is much simpler to interpret than spreadsheets full of data.
Figure 5 Injection and Production Data 2

These results can be incorporated into the 3D reservoir model to help improve history matching and prediction profiles and thus, reduces uncertainty.
Figure 6 Map of Communicating Well Pairs

0 responses | Add Yours


You must be logged in to post a comment.

We’re here to help
D Nathan Meehan