SLB, traded on the NYSE, is a technology company that works with customers to access energy. Our team, consisting of individuals from over 160 different nationalities, is dedicated to providing top-notch digital solutions and implementing cutting-edge technologies that promote performance and sustainability in the global energy sector. Our range of products and services are designed to reduce emissions and improve performance in the oil and gas industry. We are also committed to hastening the transition to low-carbon energy, and we prioritize delivering digital solutions at scale to facilitate faster, more informed decision-making. You can learn more about us atwww.slb.com
Best Acceleration Use Case
Best Moonshot Use Case
In an oil field where the primary artificial lift system is Electric Submersible Pumping (ESP), have occurred events beyond the control of the company that have resulted in the suspension of oil production operations. These events include well shutdowns as well as the occurrence of production flowline ruptures, both of which have led to the cessation of activities.
Normally in a process of oil production with Electric Submersible Pumping, the electric submersible pump lifts the fluid to surface helped by the action of the motor of the pump which is activated with electric intensity at a frequency that makes the motor spin. When the fluid is lifted, the pressure with which the fluid is entering the pump is monitored by (Pump Intake Pressure) PIP parameter.
During a shutdown period, the motor is off, and the values of electric intensity (Amps) and frequency (Hz) are equal to zero, therefore the pump does not work. However, the oil reservoir continues to supply fluid, but the ESP is no longer lifting that fluid to surface, then the fluid is not entering the pump, and that fluid is going to remain in the annular space between the production pipe and the casing of the well. Due to this condition, the fluid level in the annular space increases and PIP parameter does too. This fluid level in the annular will be so high that is going to restrict the contribution of reservoir fluids and equalize the fluid level pressure with the reservoir pressure.
This reservoir pressure is essential data for different analysis such as proposals of drilling new wells, workover operations, reservoir, and production engineering analysis, and more. When these events occur, well-by-well analysis is done to identify start and end of shutoff events and stabilized pressures to get reservoir pressure data.
However, this analysis could take at least one week to get completed, and also there is no general visualization dashboard that allow the engineer to have a general overview of the data that is being analyzed, which limit the efficiency of this process.
As a first step, the moment when a shutdown event occurs was defined using the variables of frequency, amperage, and intake pressure of the ESP. With the information of these 3 variables, the shutdown is identified giving as result the increase in pressure until it stabilizes and reaches the reservoir pressure.
To enhance data analysis and ensure quality assurance, the Production Data Foundation (PDF) plug-in is employed to extract information from the company's internal system. The retrieved data may consist of outliers, empty entries, and null values. Special attention has been given to signal processing, utilizing exploratory data analysis techniques. It is essential that these signals exhibit continuity and lack noise to ensure reliable QA/QC procedures.
By utilizing data quality techniques, we have established a robust Dataiku workflow encompassing various datasets, recipes, and Python algorithms. This workflow possesses the capability to efficiently handle substantial volumes of data from multiple wells. Its primary objective is to accurately detect pressure restoration and stabilization events within a matter of minutes.
As a result, the tool reports the start and end pressure of the shutdown with its corresponding date and time, data analysis in the continuity closure, count, and event category. The results are published in Spotfire using the Visual Analytics Spotfire BI Connector Plugin. Latitude and longitude information was added to these results for the creation of maps and analysis by areas.
The results were validated by observing and analyzing the records of stabilization pressure reported. We reached 88% success in the identification of events in a period of 5 minutes.
The technical team used to spend around 25 hours per month to get this job done, with this implementation we reduce by 76% men/hour through reservoir pressure detection Tool, achieving the automation of identification and gathering process of stabilized pressure data thanks to an efficient Dataiku workflow with a combination of datasets, recipes, and programming.
Additionally, this tool allows easy visualization of results in Spotfire using the Visual Analytics Spotfire BI Connector Plugin.
Business Area Enhanced: Analytics
Use Case Stage: Built & Functional
With this powerful analytical tool, we have optimized 76% men/hour used to analyze well by well pressure trends monthly. This generates an outstanding positive impact of the process efficiency of getting reservoir pressure data, which is essential in the different analysis carried out for daily operations in oil industry.
This impact is not just only for time efficiency during the job, but also for social life where the engineer takes advantage of the saved time for spending more time with family and friends, reducing high levels of stress.
By consistently employing this workflow, we can effectively reduce energy consumption. Considering a standard computer with an energy consumption of approximately 250 W, the utilization of the data analysis tool allows for potential savings of around 14 kWh per week during regular work hours. With this analysis being conducted every seven days, the energy savings amount to four times the weekly consumption, resulting in an impressive monthly energy conservation of 56 kWh, and an annual saving of 672 kWh.
Value Brought by Dataiku:
Dataiku has enabled this impact with easy-to-use visual interfaces to join imported datasets, group, clean, transform, and enrich data by coding. We have created a robust data analysis project, with a complete dashboard that shows the results through plots, trends and lists. Additionally, Dataiku's flow offers a distinctive collaborative environment, enabling everyone to actively contribute to the project within a shared workspace. This inclusive space allows for simultaneous collaboration, fostering teamwork and knowledge sharing among team members with diverse skill sets.
The tool has therefore enhanced speed and agility through increases team efficiency. A phase II of this project is on the way with improvements and implementation of additional resources of Dataiku that will allow to have a statistic and probability analysis of the results history.