After acquiring Berry Petroleum in 2013, Linn Energy became the operator of heavy oil fields in California’s San Joaquin Valley.
David Deaton, chief advisor for Halliburton Landmark’s North American production services, discussed challenges the company faced in optimizing the field during a technical presentation (SPE 173404) at the recent SPE Digital Energy Conference and Exhibition in The Woodlands, Texas.
“Once Linn took over [Berry], production management and surveillance issues escalated. There were a lot more wells, more complex wells, and a diverse range of asset types,” said Deaton.
Because the field required the use of steam injection, Linn had to conduct more monitoring in order to maintain maximum production efficiency.
Berry had begun a production management project, called the Reservoir Surveillance Initiative, in 2012. When Linn acquired the company, it continued the initiative.
Logistical problems that Linn sought to address were coping with the high volume of data, servicing multiple wells, and enabling engineers to make better use of their time, instead of having it consumed by data manipulation.
Using Halliburton Landmark’s DecisionSpace Platform (DSP), Linn pushed forward with the initiative and integrated disparate well and artificial lift data into one source. This helped engineers make quicker decisions and improved operations.
Data Integration
A major aspect of improving operations was visualization. Before the initiative, the data were stored in various locations without a way to access them easily for purposes such as production optimization and monitoring. Daily report data were only accessible as discrete files on a shared hard drive.
Using the platform, collaboration rooms were created with large screens, where groups of field engineers were able to see a detailed map-based display. Data once stored in spreadsheets and PDF files, which required significant manipulation, were represented by dashboards, graphs, and color coding.
The platform also enables streamlined communication with personnel working in the field. When an event occurs, it can send text and email messages alerting engineers to problems immediately after they are detected.
“The idea was to get the right data to the right people at the right time to help them make decisions,” said Deaton.
An area in which visualization gave Linn an edge was detecting and correcting for subsidence, a potentially destructive downward geological shifting.
When steam is injected into the valley’s diatomite formation, the rock dilates. After the steam ceases, and the dilation reverses, subsidence can start to occur. It can even cause casing failure, wellbore stability issues, and what Linn calls surface events.
“The diatomite will sometimes move around, open up, and create events at the surface, with steam coming out,” said Deaton.
To track the process, Linn monitors the ground around wells using tiltmeters, which detect changes in incline. The tiltmeters send data back to the platform, which overlays them onto asset maps and indicates by color the areas in which deviations from the average incline are occurring. The steam cycles can be adjusted accordingly to avoid incidents.
Before using DSP, these maps were only available as static PDF files. Now engineers can see changes as soon as new data are available and view past patterns as time-lapse animations.
Another way Linn now looks for subsidence-induced deformation is by monitoring the movement of sinker bars. The software’s dashboard enables the integration of tilt data, and production and injection data, with mechanical integrity data from the sinker bars. A deviation in sinker bar behavior may indicate casing damage, giving engineers an early warning before conditions worsen.
The initiative also helped optimize pumping for wells that use continuous steam injection instead of cyclical steam injection, which is used across the rest of the asset. In this area, the operator must maintain the optimal energy balance in the reservoir.
The Marx-Langenheim model, which calculates heat balance, was cumbersome and time-consuming to use before the initiative, so the calculations were done infrequently. However, Linn used the software to automate the process and display the output visually. As a result, the analysis may be performed immediately after new input data become available, resulting in a more efficient management of steam injection.
It took Linn approximately 1 year to work DSP into its San Joaquin operations. After fully executing the initiative, it became easier for engineers to know where their expertise is needed most, and to address small problems before they grow in scope. As a result, the company has experienced a reduced number of well and equipment failures.
Although integration, visualization, and automation enabled the initiative’s goals, the highest value can only be achieved when staff change the way they work to leverage these tools. “This [initiative] is not about the technology,” said Deaton. “This is about the value of the workflows.”