Cement evaluation data acquired in oil and gas wells for confirmation of zonal isolation, channeling in cement behind casing and well integrity. All available technologies for cement evaluations are primarily measurements of acoustic parameters like amplitude of first arrival, full waveform recording of refracted wave, impedance and attenuation or there a combination. Generally, operationsPetrophysicist, petroleum engineer or service providers are responsible for evaluation of cement bond logs and propose remedial jobs if required. This cement bond log interpretation is quite subjective if performed through pure visual interpretation and its accuracy depends on objective of well, work pressure and experience of interpreter. This paper talks about how ADNOC Onshore has leveraged machine learning (ML) for interpretation of various cement bond logs from several service providers. In green fields, cement evaluation is very important in all new wells to ensure good cement quality and zonal isolation and it is also equally important in in brown fields where 1000s of wells were drilled where well integrity issues are of common occurrence. Major challenges in evaluation are inconsistency and human bias in interpretation and as a result, interpretation may vary from one interpreter to the other. Authors have tested different ML techniques (Random Forest Classification & Neural Net) and smart way of data training. Final recommendation is to use nested models instead of single model. In this technique, input measurements and required solutions will be classified and divided into different classes, a separate ML model will be built for each class and combine all the models to get final cement evaluation and recommendations. Final results include data quality flags, cement bond quality, zonal isolation, channeling, micro annuals, areal cement map (if analyzed field wise), anomaly maps (if analyzed data from offset wells at different times), and recommendation. In this project, a simple easy to use user interface has been developed to browse the cement logs and use the trained ML models to predict the cement evaluations with a click of a button. This ML based cement bond evaluation is proved to be very effective and saved 75% of efforts by operational Petrophysicists. The interpretation accuracy has been significantly improved. This method has potential to be used at rig site. These models are very cost effective with, minimum human bias, and improves consistency as well as independent of wells or type of reservoirs. These models are tested across ADNOC Onshore Wells and can be extended to any well irrespective of geographic locations. This paper discusses machine-learning approach, evaluation of various algorithms and testing results for cement evaluation log from various service providers.

A field in Southern Oman has been identified as a potential target for Alkali-Surfactant-Polymer (ASP) flooding. Experimental investigations showed that ASP flood recovers more than 90% of waterflood remaining oil saturation. Simulation study showed that ASP increases the field recovery factor by more than 20% over waterflood or 12% over polymer flood. The potential size of the prize by ASP in this field alone is significant. The successful ASP single well chemical tracer test (SWCCT) and micro-pilot in this field validated the lab experimental results and demonstrated the significant desaturation by ASP and the extremely low oil residual after ASP. The ASP pilot was designed and initiated to de-risk and underpin the financial investment decision for a full-field implementation. The ASP pilot is designed as a small 1.4 acres (75m × 75m) inverted 5-spot pattern. It ensures completion of the field trial within one year of ASP flood. The pilot was commissioned in Q1 2014 with water pre-flush and the first ASP injection is anticipated for Q1 2016. A dynamic model was constructed and simulation was carried out. A comprehensive surveillance program was included in the pilot design and is being executed successfully. Surveillance data provide critical information to understand pattern communication, fluid flow path and reservoir characteristics. This paper will describe the results and analysis of some waterflood surveillance data and the integrated workflow of history matching. During the ASP pilot drilling campaign in 2013, an unexpected fault was encountered crossing the well pattern. This introduced an element of uncertainty regarding pattern communication. Pressure data and a passive tracer test confirmed the connectivity between the injector to 4 producers and that the fault is non-sealing. Initially the sector model was built with one fault across the pilot area, but the fluid communication in the pattern revealed by analysis of tracer suggested possible different geological realizations than the initial model. Therefore several static geological models were generated to reduce the uncertainties on history match. Well conformances were determined by PLT and DTS/DAS and were incorporated in the dynamic model. The time-lapse NMR log in the observation well provides insights on sweep, desaturation and micro displacement efficiency. Establishing and building a waterflood baseline is the foundation for the next phase of ASP implementation. This paper will share the analysis, learnings and practical implications of the pilot to date.

Majority of oil wells operated by Petroleum Development Oman (PDO) are produced by beam-pumps (BP). Average water cut in a number of fields in South of Oman reaches 95%. Increasing water production overloads processing facilities leading to handling and disposal constrains requiring wells to be shut-in. BP completions are not surveillance friendly making production logging to identify water entry for optimization (water shut-off) a challenge. The current technique to acquire production logs requires recompletion to dual-string completion to allow logging: BP short-string and surveillance conduit long-string. This is resource intensive, high cost, restricts production and limited to 9-5/8in. cased wells. Moreover, new wells are completed with dual 9-5/8in. × 7in. casing for well life-cycle integrity management. A novel solution was developed and part-funded by PDO consisting of a jet-pump (JP), 1in. inside 2in. concentric-coiled tubing (CCT) strings, power cable and production logging tools (PLT). This cost-effective real-time surveillance technique will facilitate routine production logging in BP wells, significantly reducing well intervention time and cost (50% reduction) as only the rod string is retrieved by light-hoist in preparation for logging. Wells completed with dual-string completions, which have previously been production logged were selected for field trial. These existing logs were used as a baseline for new log comparison. The technique was successfully deployed in a 3 well field trial campaign for the first time in southern oilfields (industry first). The new production logs compared very well to existing logs (same water signature observed), proving the techniques robustness to identify water entry in different production environments. We preset advantages of the new technique over conventional, candidate selection, logging tool options, interpretation methodology, field trial results and comparison logs. This new system is being deployed across PDO and is applicable to other fields being produced by BP, progressing-cavity pump (PCP) or electrical submersible pump (ESP) to identify water entry for production enhancement or reservoir monitoring.

Enhanced Oil Recovery (EOR) processes are key to Petroleum Development Oman (PDO) longer term business performance. To date, PDO is operating four commercial scale EOR projects and a number of pilots that are either ongoing or recently concluded. The EOR projects and pilots cover chemical and thermal EOR as well as miscible gas injection. Successful EOR projects require robust long term strategic plans with built-in flexibility and seamless execution, in order to continuously de-risk associated uncertainties through proper testing and piloting. One of the key contributors to PDO's successful EOR journey has been successful monitoring and surveillance through acquisition of high quality surveillance data. An alkaline surfactant polymer (ASP) pilot, first of its kind in PDO was recently concluded with encouraging results. Key pilot successes parameters included achieving reduction of oil saturation to less than 10% in one layer at the observation well. The challenge of saturation monitoring through salinity independent and carbon insensitive technique in EOR fields was addressed by Nuclear Magnetic Resonance (NMR) time-lapse cased hole logging through fiber-reinforced plastic (FRP) casing. The other ongoing EOR pilot involves injecting polymer into a heavy oil bearing reservoir with a strong bottom aquifer drive. In this pilot, the key subsurface uncertainties are polymer injectivity, conformance and sweep efficiency. These uncertainties are being de-risked by deploying monitoring technologies such as distributed temperature sensing (DTS), distributed acoustic sensing (DAS), Pressure monitoring and time-lapse saturation logging based on both nuclear and electrical principals. Some of the challenges in the miscible gas injection project include; gas breakthrough evaluation, reservoir connectivity, and gas sweep efficiency. These were assessed by implementing inter-well tracer test, production and time-lapse saturation loggings. Surveillance in Thermal EOR project (cyclic steam soak, CSS) revolves around having dedicated temperature, pressure observation wells and systematic temperature surveillance across the field. Assessment of steam injection profile and steam quality has also been focus areas. The aim is not only to monitor areal and vertical sweep efficiency (of both steam and reservoir fluids) over time, but also to get leading signals for a proper reservoir management to maximize profitability. Further, pattern recognition from microseismic survey data helps monitoring the ‘cap-rock’ integrity and reservoir containment. Production logging in the ultra high viscosity oil zone still remains a challenge. Detailed fiber optics reservoir monitoring was implemented in Thermal EOR in naturally fractured carbonate reservoir. The objectivities are the oil rim-management and safeguard the cap-rock integrity from fault re-activation In this paper, PDO's experiences in handling EOR challenges and how different EOR monitoring and surveillance technologies were utilized will be presented. A recommended practice will be discussed based on PDO's experience.

Wettability is a critical reservoir and petrophysical evaluation parameter that is often ignored. Both disciplines often assume the formations are water-wet for simplicity and because wettability measurement on cores often carries a high degree of uncertainty. With the expansion of unconventional carbonate reservoirs development and interest in enhanced oil recovery (EOR), the importance of understanding wettability at the native state and its variability with various injection fluids is becoming critical. For practical purposes, a fast and accurate determination method, ideally at in-situ conditions, is desired. It is widely recognized that nuclear magnetic resonance (NMR) is very sensitive to the strength of the fluid-rock interactions, and therefore, has been long considered as a good candidate for wettability determination. The NMR methodology was first applied in the laboratory using T2 relaxation measurements. For instance, sample wettability is inferred from a shift of the oil peak to shorter T2 values compared with the bulk T2 response of a live oil in the case of oil-wet system. The main practical limitation to the applicability of the T2 shift-based evaluation of wettability is the usually poor separation of oil and water peaks in the T2 spectrum. Furthermore, the bulk T2 of live oils must be measured and the core sample must be perfectly cleaned to quantify the NMR surface relaxation effect. Recently, a method based on two-dimensional mapping of NMR diffusion versus T2 was developed and validated with Amott-Harvey and USBM lab measurements. This method has two advantages. First, separation between the oil and water signals is greatly improved compared with the one-dimensional T2. Second, key properties such as tortuosity, represented by the electrical cementation factor m, and effective surface relaxivity can be inferred from the two-dimensional NMR maps using the restricted-diffusion model. The wettability index can then be estimated from the effective surface relaxivities. The laboratory results on cores suggest that it is possible to obtain reservoir wettability using downhole NMR measurements. This requires high-resolution, high signal-to-noise ratio (SNR) data and improved processing techniques to separate oil and water signals. We examined the NMR restricted-diffusion wettability technique utilizing log data collected in an observation well completed with plastic casing. This well is used to monitor oil desaturation during different phases of an EOR pilot consisting of water, alkaline surfactant (ASP), and polymer floods. A downhole NMR tool that simultaneously records T1, T2 and diffusion at multiple depth of investigation (DOI) was used. This device allowed to periodically collect high-quality NMR data with SNR higher than 50. The targeted reservoir is a sandstone containing hydrocarbon with viscosity of 90 cP. The computed wettability consistently showed mildly oil-wet condition at the selected depth and over the analyzed time intervals.