BAYESIAN METHOD FOR RAPID MULTI-WELL INTERPRETATION OF WELL LOGS AND CORE DATA IN UNCONVENTIONAL FORMATIONS

2020 ◽  
Author(s):  
Tianqi Deng ◽  
◽  
Joaquín Ambía ◽  
Carlos Torres-Verdín ◽  
◽  
...  
Keyword(s):  
2007 ◽  
Vol 10 (06) ◽  
pp. 711-729 ◽  
Author(s):  
Paul Francis Worthington

Summary A user-friendly type chart has been constructed as an aid to the evaluation of water saturation from well logs. It provides a basis for the inter-reservoir comparison of electrical character in terms of adherence to, or departures from, Archie conditions in the presence of significant shaliness and/or low formation-water salinity. Therefore, it constitutes an analog facility. The deliverables include reservoir classification to guide well-log analysis, a protocol for optimizing the acquisition of special core data in support of log analysis, and reservoir characterization in terms of an (analog) porosity exponent and saturation exponent. The type chart describes a continuum of electrical behavior for both water and hydrocarbon zones. This is important because some reservoir rocks can conform to Archie conditions in the fully water-saturated state, but show pronounced departures from Archie conditions in the partially water-saturated state. In this respect, the chart is an extension of earlier approaches that were restricted to the water zone. This extension is achieved by adopting a generalized geometric factor—the ratio of water conductivity to formation conductivity—regardless of the degree of hydrocarbon saturation. The type chart relates a normalized form of this geometric factor to formation-water conductivity, a "shale" conductivity term, and (irreducible) water saturation. The chart has been validated using core data from comprehensively studied reservoirs. A workflow details the application of the type chart to core and/or log data. The analog role of the chart is illustrated for reservoir units that show different levels of non-Archie effects. The application of the method should take rock types, scale effects, the degree of core sampling, and net reservoir criteria into account. The principal benefit is a reduced uncertainty in the choice of a procedure for the petrophysical evaluation of water saturation, especially at an early stage in the appraisal/development process, when adequate characterizing data may not be available. Introduction One of the ever-present problems in petrophysics is how to carry out a meaningful evaluation of well logs in situations where characterizing information from quality-assured core analysis is either unavailable or is insufficient to satisfactorily support the log interpretation. This problem is especially pertinent at an early stage in the life of a field, when reservoir data are relatively sparse. Data shortfalls could be mitigated if there was a means of identifying petrophysical analogs of reservoir character, so that the broader experience of the hydrocarbon industry could be utilized in constructing reservoir models and thence be brought to bear on current appraisal and development decisions. Here, a principal requirement calls for type charts of petrophysical character, on which data from different reservoirs can be plotted and compared, as a basis for aligning approaches to future data acquisition and interpretation. This need manifests itself strongly in the petrophysical evaluation of water saturation, a process that traditionally uses the electrical properties of a reservoir rock to deliver key building blocks for an integrated reservoir model. The solution to this problem calls for an analog facility through which the electrical character of a subject reservoir can be compared with others that have been more comprehensively studied. In this way, the degree of confidence in log-derived water saturation might be reinforced. At the limit, the log analyst needs a reference basis for recourse to capillary pressure data in cases where the well-log evaluation of water saturation turns out to be prohibitively uncertain.


2017 ◽  
Vol 5 (2) ◽  
pp. 57 ◽  
Author(s):  
Godwin Aigbadon ◽  
A.U Okoro ◽  
Chuku Una ◽  
Ocheli Azuka

The 3-D depositional environment was built using seismic data. The depositional facies was used to locate channels with highly theif zones and distribution of various sedimentary facies. The integration core data and the gamma ray log trend from the wells within the studied interval with right boxcar/right bow-shape indicate muddy tidal flat to mixed tidal flat environments. The bell–shaped from the well logs with the core data indicate delta front with mouth bar, the blocky box- car trend from the well logs with the core data indicate tidal point bar with tidal channel fill. The integration of seismic to well log tie display a good tie in the wells across the field. The attribute map from velocity analysis revealed the presence of hydrocarbons in the identified sands (A, B, C, D1, D2, D4, D5). The major faults F1, F2, F3 and F4 with good sealing capacity are responsible for hydrocarbon accumulation in the field. Detailed petro physical analysis of well log data showed that the studied interval are characterized by sand-shale inter-beds. Eight reservoirs were mapped at depth intervals of 2886m to 3533m with their thicknesses ranging from 12m to 407m. Also the Analysis of the petrophysical results showed that porosity of the reservoirs range from 14% to 28 %; permeability range from 245.70 md to 454.7md; water saturation values from 21.65% to 54.50% and hydrocarbon saturation values from 45.50% to 78.50 %. The by-passed hydrocarbons were identified and estimated in low resistivity pay sands D1, D2 at depth of 2884m – 2940m, sand D5 at depth of 3114m – 3126m respectively. The model serve as a basis for establishing facies model in the field.


2019 ◽  
Vol 7 (1) ◽  
pp. 58
Author(s):  
G. O. Aigbadon ◽  
E. O. Akpunonu ◽  
S. O. Agunloye ◽  
A. Ocheli ◽  
O. O .Akakaru

This study was carried out integrating well logs and core to build reservoir model for the Useni-1 oil field. Core data and well logs were used to evaluate the petrophysical characteristics of the reservoirs. The paleodepositional environment was deduce from the wells and cores data. The depositional facies model showed highly permeable channels where the wells where positioned. The environments identified that the fluvial channel facies with highly permeable zones constituted the reservoirs. Four reservoirs were mapped at depth range of 8000ft to 8400ft with thicknesses varying from 20ft to 400ft. Petrophysical results showed that porosity of the reservoirs varied from 12% to 28 %; permeability from 145.70 md to 454.70md; water saturation from 21.65% to 54.50% and hydrocarbon saturation from 45.50% to 78.50 %. Core data and the gamma ray log trends with right boxcar trend indicate fluvial point bar and tidal channel fills in the lower delta plain setting. By-passed hydrocarbons were identified in low resistivity pay sands D1, D2 at depth of 7800 – 78100ft in the field.  


2016 ◽  
Vol 4 (2) ◽  
pp. SF19-SF29
Author(s):  
Chicheng Xu ◽  
Qinshan Yang ◽  
Carlos Torres-Verdín

Rock typing is critical in deepwater reservoir characterization to construct stratigraphic models populated with static and dynamic petrophysical properties. Rock typing based on multiple well logs is subject to large uncertainty in thinly bedded reservoirs because true physical properties cannot be resolved by low-resolution logging tools due to shoulder-bed effects. We have introduced a new Bayesian approach that inherently adopts the scientific method of iterative hypothesis testing to perform rock typing by simultaneously honoring different logging-tool physics in a multilayered earth model. In addition to estimating the vertical distribution of rock types with maximum likelihood, the Bayesian method quantifies the uncertainty of rock types and the associated petrophysical properties layer by layer. Bayesian rock classification is performed with a fast sampling technique based on the Markov-chain Monte Carlo method, thereby enabling an efficient search of rock types to obtain the final results. We have used a fast linear iterative refinement method to simulate nuclear logs and a 2D forward modeling code to simulate array-induction resistivity logs. A rock-type distribution hypothesis is considered acceptable only when all the observed well logs are reproduced with forward modeling. In a field case of offshore deltaic gas reservoir, the Bayesian method differentiates rock types that exhibit subtle petrophysical variations due to grain size change. The new method provides more than 77% agreement between log- and core-derived rock types, whereas conventional deterministic methods achieve only 60% agreement due to the presence of thin beds and laminations. Even though large uncertainty is observed in thinly bedded and laminated zones, the Bayesian rock-typing method still yields rock types and petrophysical properties that agree well with core-plug measurements acquired in these layers. As a result, the overall correlation between log-derived permeability and core-measured permeability is improved by approximately 16% when compared with conventional deterministic methods.


Geophysics ◽  
2003 ◽  
Vol 68 (6) ◽  
pp. 2000-2009 ◽  
Author(s):  
Arild Buland ◽  
Henning Omre

A Bayesian method for wavelet estimation from seismic and well data is developed. The method works both on stacked data and on prestack data in form of angle gathers. The seismic forward model is based on the convolutional model, where the reflectivity is calculated from the well logs. Possible misties between the seismic traveltimes and the time axis of the well logs, errors in the log measurements, and seismic noise are included in the model. The estimated wavelets are given as probability density functions such that uncertainties of the wavelets are an integral part of the solution. The solution is not analytically obtainable and is therefore computed by Markov‐chain Monte Carlo simulation. An example from Sleipner field shows that the estimated wavelet has higher amplitude compared to wavelet estimation where well log errors are neglected, and the uncertainty of the estimated wavelet is lower.


2015 ◽  
Vol 3 (1) ◽  
pp. SAi-SAii
Author(s):  
Chicheng Xu ◽  
Carlos Torres-Verdín ◽  
Kyle Spikes ◽  
Thaimar Ramirez
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document