Geophysical data allows for measuring a change in petrophysical parameters thought a whole well length. They often exhibit a chaotic behaviour which is difficult to describe and finding a pattern is near impossible. A potential measure of this chaos – correlation dimension – has been examined in the study. The research was carried out for the log data from Williston Basin, USA and the Norwegian Lille-Frigg oil field on the North Sea. Sonic log (DT), neutron porosity log (NPHI), deep resistivity log (LLD) as well as density log (RHOB) were utilised in the study. A python program has been written to measure the change in correlation dimension. Instead of calculating a one value of a correlation dimension for a whole log, a moving range algorithm was developed and implemented. It is based on defining a range for which the dimension is calculated and then moving the range on a geophysical log. In addition, a graph representing change of a correlation dimension with depth is drawn. The influence of data range and range shift were measured. Over 100 correlations have been carried out between rock properties and their dimension. The results indicate that the correlation dimensions change throughout the whole geophysical log and correlate with themselves and other curves in a moderate degree. It allows for determining ranges where a data set is not chaotic. The research shows that properly set range should have a reasonable and representative amount of data points, while the shift should be small for accurate results. Presented analysis creates perspectives for a more precise rock formation description and possible correlation between different oil wells within a single reservoir.
{"title":"Randomness of Geophysical Log Data – Fractal Approach","authors":"M. Figiel","doi":"10.2118/199776-stu","DOIUrl":"https://doi.org/10.2118/199776-stu","url":null,"abstract":"\u0000 Geophysical data allows for measuring a change in petrophysical parameters thought a whole well length. They often exhibit a chaotic behaviour which is difficult to describe and finding a pattern is near impossible. A potential measure of this chaos – correlation dimension – has been examined in the study.\u0000 The research was carried out for the log data from Williston Basin, USA and the Norwegian Lille-Frigg oil field on the North Sea. Sonic log (DT), neutron porosity log (NPHI), deep resistivity log (LLD) as well as density log (RHOB) were utilised in the study. A python program has been written to measure the change in correlation dimension. Instead of calculating a one value of a correlation dimension for a whole log, a moving range algorithm was developed and implemented. It is based on defining a range for which the dimension is calculated and then moving the range on a geophysical log. In addition, a graph representing change of a correlation dimension with depth is drawn. The influence of data range and range shift were measured. Over 100 correlations have been carried out between rock properties and their dimension.\u0000 The results indicate that the correlation dimensions change throughout the whole geophysical log and correlate with themselves and other curves in a moderate degree. It allows for determining ranges where a data set is not chaotic. The research shows that properly set range should have a reasonable and representative amount of data points, while the shift should be small for accurate results. Presented analysis creates perspectives for a more precise rock formation description and possible correlation between different oil wells within a single reservoir.","PeriodicalId":10909,"journal":{"name":"Day 2 Tue, October 01, 2019","volume":"69 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91388366","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Giuseppe Feo, Jyotsna Sharma, W. Williams, Dmitry Kortukov, O. Toba
Effective well control depends on the drilling teams’ knowledge of wellbore flow dynamics and their ability to predict and control influx. Detection of a gas influx in an offshore environment is particularly challenging, and there are no existing datasets that have been verified and validated for gas kick migration at full scale annulus conditions. This study bridges this gap with the newly instrumented experimental well at PERTT (Petroleum Engineering Research & Technology Transfer Lab) at Louisiana State University (LSU) simulating an offshore marine riser environment with its larger than average annular space and mud circulation capability. The experimental setup instrumented with fiber optics and pressure/temperature gauges provides a physical model of the dynamic gas migration over large distances in full scale annular conditions. Current kick detection methods do not always reliably detect a gas influx and have not kept pace with the increasingly challenging offshore drilling conditions. Even though there have been some recent developments in offshore kick detection, all methods thus far are only qualitative in nature because they are based on measurements at the surface. This study addresses current kick detection limitations and illuminates the potential for implementing distributed fiber optic sensing (DFOS) to the marine riser as a non-invasive and effective kick detection method in both stagnant and circulating annular conditions. As North America's only academic full scale well testing center, an experimental well in the PERTT lab was utilized to monitor and characterize gas rise using DFOS to simulate well control scenarios in offshore drilling riser environments. DFOS allows for the tracking of the gas migration in both the stagnant and full-scale circulating annulus conditions. Data from pressure sensors is integrated with the distributed temperature (DTS) and acoustic (DAS) measurements, for real-time downhole monitoring of the dynamics of the gas migration and fluid front movement. By implementing time and frequency domain analysis of the fiber optic data, we show that the gas rise and water front movement can be identified. Both the water and gas injection down the tubing independently show characteristic fronts in the DTS and DAS data, which gives us confidence in our interpretation. Once the gas is present in the annulus, the DAS measurements indicate a higher than expected gas-rise velocity, and this is most probably due to the full-scale annular geometry and circulating conditions enabling a faster gas rise velocity compared to previous work in this area consisting only of small-scale experiments and experiments through tubing. The two-phase flow experiments conducted in this research provide critical insights for understanding the flow dynamics in offshore drilling riser conditions, and the results provide an indication of how quickly gas can migrate in a marine riser scenario warranting further investigation for the sak
{"title":"Application of Distributed Fiber Optics Sensing Technology for Real-Time Gas Kick Detection","authors":"Giuseppe Feo, Jyotsna Sharma, W. Williams, Dmitry Kortukov, O. Toba","doi":"10.2118/196113-ms","DOIUrl":"https://doi.org/10.2118/196113-ms","url":null,"abstract":"\u0000 Effective well control depends on the drilling teams’ knowledge of wellbore flow dynamics and their ability to predict and control influx. Detection of a gas influx in an offshore environment is particularly challenging, and there are no existing datasets that have been verified and validated for gas kick migration at full scale annulus conditions. This study bridges this gap with the newly instrumented experimental well at PERTT (Petroleum Engineering Research & Technology Transfer Lab) at Louisiana State University (LSU) simulating an offshore marine riser environment with its larger than average annular space and mud circulation capability. The experimental setup instrumented with fiber optics and pressure/temperature gauges provides a physical model of the dynamic gas migration over large distances in full scale annular conditions. Current kick detection methods do not always reliably detect a gas influx and have not kept pace with the increasingly challenging offshore drilling conditions. Even though there have been some recent developments in offshore kick detection, all methods thus far are only qualitative in nature because they are based on measurements at the surface. This study addresses current kick detection limitations and illuminates the potential for implementing distributed fiber optic sensing (DFOS) to the marine riser as a non-invasive and effective kick detection method in both stagnant and circulating annular conditions. As North America's only academic full scale well testing center, an experimental well in the PERTT lab was utilized to monitor and characterize gas rise using DFOS to simulate well control scenarios in offshore drilling riser environments.\u0000 DFOS allows for the tracking of the gas migration in both the stagnant and full-scale circulating annulus conditions. Data from pressure sensors is integrated with the distributed temperature (DTS) and acoustic (DAS) measurements, for real-time downhole monitoring of the dynamics of the gas migration and fluid front movement. By implementing time and frequency domain analysis of the fiber optic data, we show that the gas rise and water front movement can be identified. Both the water and gas injection down the tubing independently show characteristic fronts in the DTS and DAS data, which gives us confidence in our interpretation. Once the gas is present in the annulus, the DAS measurements indicate a higher than expected gas-rise velocity, and this is most probably due to the full-scale annular geometry and circulating conditions enabling a faster gas rise velocity compared to previous work in this area consisting only of small-scale experiments and experiments through tubing. The two-phase flow experiments conducted in this research provide critical insights for understanding the flow dynamics in offshore drilling riser conditions, and the results provide an indication of how quickly gas can migrate in a marine riser scenario warranting further investigation for the sak","PeriodicalId":10909,"journal":{"name":"Day 2 Tue, October 01, 2019","volume":"121 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73055896","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Caja, A. Peña, J. R. Campos, Laura García Diego, J. Tritlla, T. Bover‐Arnal, J. Martín‐Martín
Cuttings provide the opportunity to precisely look at the rock that has been drilled. A preliminary drill cuttings description is commontly performed by mudloggers and wellsite geologists using conventional binocular microscope at the drilling rig. After this preliminary description, often the bags of cuttings are stored in a warehouse and samples are seldom examined back again. Cuttings give the geologist information about the formation lithology needed for geologic correlation, understanding about reservoir quality, seals and source rocks, and can also be an input for the petrophysicist. In this study, we are testing a methodology to identify, classify and quantify lithologies present in cutting samples using thin section images. The method includes sample preparation (washing, drying and thin section cuttings preparation), image acquisition (to obtain whole thin section gigapixel high resolution microscopy images), virtual microscopy (to identify lithologies) and automatic image analysis (to perform supervised machine learning lithology clasiffication). Virtual microscopy allowed the identification of four main lithologies in all the studied thin sections: quartzites (including loose quartz grains), siltstones, claystones and carbonates. Image analysis allowed the classification and quantification of the identified lithologies in 16 drill cutting samples from two tight gas reservoirs. This innovative methodology allowed the fast identification of lithologies using virtual microscopy and their classification and quantification by image analysis and supervised machine learning. This approach is widely accessible as open source software was used for virtual microscopy and image analysis. Algorithm training and model generation was relativelly fast, and its performance or accuracy was qualititavely evaluated by virtual microscopy with good classification results.
{"title":"Image Processing and Machine Learning Applied to Lithology Identification, Classification and Quantification of Thin Section Cutting Samples","authors":"M. Caja, A. Peña, J. R. Campos, Laura García Diego, J. Tritlla, T. Bover‐Arnal, J. Martín‐Martín","doi":"10.2118/196117-ms","DOIUrl":"https://doi.org/10.2118/196117-ms","url":null,"abstract":"\u0000 Cuttings provide the opportunity to precisely look at the rock that has been drilled. A preliminary drill cuttings description is commontly performed by mudloggers and wellsite geologists using conventional binocular microscope at the drilling rig. After this preliminary description, often the bags of cuttings are stored in a warehouse and samples are seldom examined back again. Cuttings give the geologist information about the formation lithology needed for geologic correlation, understanding about reservoir quality, seals and source rocks, and can also be an input for the petrophysicist.\u0000 In this study, we are testing a methodology to identify, classify and quantify lithologies present in cutting samples using thin section images. The method includes sample preparation (washing, drying and thin section cuttings preparation), image acquisition (to obtain whole thin section gigapixel high resolution microscopy images), virtual microscopy (to identify lithologies) and automatic image analysis (to perform supervised machine learning lithology clasiffication).\u0000 Virtual microscopy allowed the identification of four main lithologies in all the studied thin sections: quartzites (including loose quartz grains), siltstones, claystones and carbonates. Image analysis allowed the classification and quantification of the identified lithologies in 16 drill cutting samples from two tight gas reservoirs.\u0000 This innovative methodology allowed the fast identification of lithologies using virtual microscopy and their classification and quantification by image analysis and supervised machine learning. This approach is widely accessible as open source software was used for virtual microscopy and image analysis. Algorithm training and model generation was relativelly fast, and its performance or accuracy was qualititavely evaluated by virtual microscopy with good classification results.","PeriodicalId":10909,"journal":{"name":"Day 2 Tue, October 01, 2019","volume":"37 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73675690","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Andrew Cunningham, N. Wentzel, T. Knode, Tony Pooley
In order to improve HSE performance many companies have implemented voluntary (i.e. non-regulation driven) programs designed to engage supervisors and employees and reduce injuries and incidents. Over the years these programs have had significant effect in improving performance and making the workplace safer. While done with the best intentions, most programs introduce an element of administrative burden on the organization and sites. The cumulative impact on a supervisor's daily activities can be substantial and result in excessive time spent in front of a computer, rather than with their teams. This means less opportunity to provide leadership on safety and consequently, undermine efforts to improve. In 2017, based on a combination of employee surveys and safety stand downs Dyno Nobel North America (‘DNA’ or the company), a global explosives manufacturer and service provider, identified the need to evaluate the burden on the organization of safety programs to rationalize and improve them as appropriate. One of the main concerns of this effort was how to remove or modify these programs to be less of a burden, yet not increase the risk. It can be related to the game Jenga®, where players remove blocks from a stack without destabilizing the structure. DNA engaged a consultant, The Jonah Group, to build a risk model based on the principles of process safety management interwoven with the understanding of human factors and performance. Once the model was built, it was piloted at three of the company's field sites to ensure efficacy and adjust as necessary. Afterwards, the model was used at nine field locations. The evaluation included a review of equipment, process and procedure, and centered around interviews with supervisors and front-line employees. Surveys were conducted with supervisors to complete the view of where they spend their time. Results and recommendations were summarized in a report. One of the key findings was that while there were opportunities to improve certain elements of the voluntary safety programs, there were more significant opportunities with regards to management of change, process safety and risk awareness, site safety leadership, communication, and process efficiency. The recommendations will help the company improve organizational effectiveness and free up supervisors to better oversee, and lead, site safety.
为了提高HSE绩效，许多公司实施了自愿(即非监管驱动)计划，旨在让主管和员工参与进来，减少伤害和事故。多年来，这些项目在提高绩效和使工作场所更安全方面产生了重大影响。虽然有最好的意图，但大多数程序会给组织和站点带来管理负担。对主管日常活动的累积影响可能是巨大的，导致他们花在电脑前的时间过多，而不是与他们的团队在一起。这意味着在安全方面发挥领导作用的机会减少，从而破坏了改进工作的努力。2017年，全球爆炸物制造商和服务提供商Dyno Nobel North America(“DNA”或公司)根据员工调查和安全状况，确定有必要评估安全计划组织的负担，以适当地合理化和改进它们。这项工作的主要关注点之一是如何删除或修改这些程序以减轻负担，同时不增加风险。这可能与叠叠乐(Jenga®)游戏有关，玩家可以在不破坏结构的情况下从堆叠中移除块。DNA聘请了咨询公司乔纳集团(The Jonah Group)，在过程安全管理原则的基础上，与对人为因素和绩效的理解相互交织，建立了一个风险模型。一旦模型建立，它将在公司的三个现场进行试验，以确保效果并根据需要进行调整。随后，该模型在9个野外地点进行了应用。评估包括对设备、流程和程序的审查，并以对主管和一线员工的采访为中心。调查是与主管一起进行的，以完成他们在哪里花费时间的看法。结果和建议总结在一份报告中。其中一个重要的发现是，虽然有机会改进自愿安全计划的某些要素，但在变革管理、过程安全和风险意识、现场安全领导、沟通和过程效率方面，有更重要的机会。这些建议将有助于该公司提高组织效率，并释放监管人员，以便更好地监督和领导现场安全。
{"title":"A Risk-Based Approach to Evaluating, and Rationalizing, the Portfolio of Company HSE Programs","authors":"Andrew Cunningham, N. Wentzel, T. Knode, Tony Pooley","doi":"10.2118/196052-ms","DOIUrl":"https://doi.org/10.2118/196052-ms","url":null,"abstract":"\u0000 In order to improve HSE performance many companies have implemented voluntary (i.e. non-regulation driven) programs designed to engage supervisors and employees and reduce injuries and incidents. Over the years these programs have had significant effect in improving performance and making the workplace safer. While done with the best intentions, most programs introduce an element of administrative burden on the organization and sites. The cumulative impact on a supervisor's daily activities can be substantial and result in excessive time spent in front of a computer, rather than with their teams. This means less opportunity to provide leadership on safety and consequently, undermine efforts to improve.\u0000 In 2017, based on a combination of employee surveys and safety stand downs Dyno Nobel North America (‘DNA’ or the company), a global explosives manufacturer and service provider, identified the need to evaluate the burden on the organization of safety programs to rationalize and improve them as appropriate. One of the main concerns of this effort was how to remove or modify these programs to be less of a burden, yet not increase the risk. It can be related to the game Jenga®, where players remove blocks from a stack without destabilizing the structure.\u0000 DNA engaged a consultant, The Jonah Group, to build a risk model based on the principles of process safety management interwoven with the understanding of human factors and performance. Once the model was built, it was piloted at three of the company's field sites to ensure efficacy and adjust as necessary. Afterwards, the model was used at nine field locations. The evaluation included a review of equipment, process and procedure, and centered around interviews with supervisors and front-line employees. Surveys were conducted with supervisors to complete the view of where they spend their time.\u0000 Results and recommendations were summarized in a report. One of the key findings was that while there were opportunities to improve certain elements of the voluntary safety programs, there were more significant opportunities with regards to management of change, process safety and risk awareness, site safety leadership, communication, and process efficiency. The recommendations will help the company improve organizational effectiveness and free up supervisors to better oversee, and lead, site safety.","PeriodicalId":10909,"journal":{"name":"Day 2 Tue, October 01, 2019","volume":"39 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73737734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Many investigations have been discussed and it is a well-recognized fact that sonic wave velocity is not only influenced by its rock matrix and the fluids occupying the pores but also by the pore architecture details of the rock bulk. This situation still brings a lack of understanding, and this study is purposed to clearly explain how acoustic velocity and quality factor correlate with porosity, permeability and details internal pore structure in porous rocks. This study employs 67 sandstone and 120 carbonate core samples collected from several countries in Europe, Australia, Asia, and USA. The measured values are available for porosity ϕ, permeability k, clay content Vcl, compressional velocity Vp, and quality factor Qp in saturated and pressurized conditions. Then, a proposed method is developed by re-arrangement on Kozeny equation to perform rock typing based on pore structure similarity which called as pore geometry-structure (PGS). The proposed rock typing method allows investigating the influential primary factors that control acoustic velocity and quality factor. Besides that, basic rock physics equations for sonic velocity and critical porosity concepts are also involved and derived to obtain a new solution to predict porosity and permeability. At least eight rock groups are established from rock typing with its Kozeny constant. This constant is a product of pore shape factor Fs and tortuosity τ. Then, the relations of velocity and quality factor versus porosity, permeability, pore geometry (k/ϕ)0.5, and pore structure (k/ϕ3) are constructed. One can find that each relation among the rock groups of each lithology is clearly separated and produce high correlations. Velocity and quality factor tend to be high with an increase in Kozeny constant. However, for a given porosity for all the groups, velocity and quality factor increase remarkably with a decrease in Kozeny constant. These all mean that velocity and quality factor increase with either an increase in the complexity of pore systems or, at the same pore complexity, a decrease in specific internal surface area. On the other hand, each rock group for both sandstone and carbonate has its critical porosity and it strongly correlates with velocity and porosity. Finally, critical porosity becomes a specific property of rock groups having similar pore geometry and structure. As a novelty, the empirical equations are derived to estimate compressional velocity and quality factor based on petrophysical parameters. Furthermore, this study also establishes empirical equations for predicting porosity and permeability by using compressional wave velocity, critical porosity, and PGS rock typing.
{"title":"New Approaches of Porosity-Permeability Estimations and Quality Factor Q Characterization based on Sonic Velocity, Critical Porosity, and Rock Typing","authors":"M. Akbar","doi":"10.2118/199777-stu","DOIUrl":"https://doi.org/10.2118/199777-stu","url":null,"abstract":"\u0000 Many investigations have been discussed and it is a well-recognized fact that sonic wave velocity is not only influenced by its rock matrix and the fluids occupying the pores but also by the pore architecture details of the rock bulk. This situation still brings a lack of understanding, and this study is purposed to clearly explain how acoustic velocity and quality factor correlate with porosity, permeability and details internal pore structure in porous rocks.\u0000 This study employs 67 sandstone and 120 carbonate core samples collected from several countries in Europe, Australia, Asia, and USA. The measured values are available for porosity ϕ, permeability k, clay content Vcl, compressional velocity Vp, and quality factor Qp in saturated and pressurized conditions. Then, a proposed method is developed by re-arrangement on Kozeny equation to perform rock typing based on pore structure similarity which called as pore geometry-structure (PGS). The proposed rock typing method allows investigating the influential primary factors that control acoustic velocity and quality factor. Besides that, basic rock physics equations for sonic velocity and critical porosity concepts are also involved and derived to obtain a new solution to predict porosity and permeability.\u0000 At least eight rock groups are established from rock typing with its Kozeny constant. This constant is a product of pore shape factor Fs and tortuosity τ. Then, the relations of velocity and quality factor versus porosity, permeability, pore geometry (k/ϕ)0.5, and pore structure (k/ϕ3) are constructed. One can find that each relation among the rock groups of each lithology is clearly separated and produce high correlations. Velocity and quality factor tend to be high with an increase in Kozeny constant. However, for a given porosity for all the groups, velocity and quality factor increase remarkably with a decrease in Kozeny constant. These all mean that velocity and quality factor increase with either an increase in the complexity of pore systems or, at the same pore complexity, a decrease in specific internal surface area. On the other hand, each rock group for both sandstone and carbonate has its critical porosity and it strongly correlates with velocity and porosity. Finally, critical porosity becomes a specific property of rock groups having similar pore geometry and structure.\u0000 As a novelty, the empirical equations are derived to estimate compressional velocity and quality factor based on petrophysical parameters. Furthermore, this study also establishes empirical equations for predicting porosity and permeability by using compressional wave velocity, critical porosity, and PGS rock typing.","PeriodicalId":10909,"journal":{"name":"Day 2 Tue, October 01, 2019","volume":"199 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78905755","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Carbonate rocks have a very complex pore system due to the presence of interparticle and intra-particle porosities. This makes the acquisition and analysis of the petrophysical data, and the characterization of carbonate rocks a big challenge. In this study, functional network tool is used to develop a model to predict water saturation using petrophysical well logs as input data and the dean-stark measured water saturation as an output parameter. The data comprised of more than 200 well log points corresponding to available core data. The developed FN model was optimized by using several optimization algorithms such as differential evolution (DE), particle swarm optimization (PSO), and covariance matrix adaptation evolution strategy (CMAES). FN model optimized with PSO found to be the most robust artificial intelligence (AI) model to predict water saturation in carbonate rocks. The results showed that the proposed model predicted the water saturation with an accuracy of 97% when related to the experimental core values. In this study in addition to the development of optimized FN model, an explicit empirical correlation is also extracted from the optimized FN model. To validate the proposed correlation, three most commonly applied water saturation models (Simandoux, Bardon and Pied model, Fertl and Hammack Model, Waxman-Smits, and Indonesian) from literature were selected and subjected to same well log data as the AI model to estimate water saturation. The estimated water saturation values for AI and other saturation models were then compared with experimental values of testing data and the results showed that AI model was able to predict water saturation with an error of less than 5% while the saturation models did the same with lesser accuracy of error up to 50%. This work clearly shows that computer-based machine learning techniques can determine water saturation with a high precision and the developed correlation works extremely well in prediction mode.
{"title":"An Artificial Intelligence Approach to Predict the Water Saturation in Carbonate Reservoir Rocks","authors":"Zeeshan Tariq, M. Mahmoud, A. Abdulraheem","doi":"10.2118/195804-ms","DOIUrl":"https://doi.org/10.2118/195804-ms","url":null,"abstract":"\u0000 Carbonate rocks have a very complex pore system due to the presence of interparticle and intra-particle porosities. This makes the acquisition and analysis of the petrophysical data, and the characterization of carbonate rocks a big challenge. In this study, functional network tool is used to develop a model to predict water saturation using petrophysical well logs as input data and the dean-stark measured water saturation as an output parameter. The data comprised of more than 200 well log points corresponding to available core data. The developed FN model was optimized by using several optimization algorithms such as differential evolution (DE), particle swarm optimization (PSO), and covariance matrix adaptation evolution strategy (CMAES). FN model optimized with PSO found to be the most robust artificial intelligence (AI) model to predict water saturation in carbonate rocks. The results showed that the proposed model predicted the water saturation with an accuracy of 97% when related to the experimental core values. In this study in addition to the development of optimized FN model, an explicit empirical correlation is also extracted from the optimized FN model. To validate the proposed correlation, three most commonly applied water saturation models (Simandoux, Bardon and Pied model, Fertl and Hammack Model, Waxman-Smits, and Indonesian) from literature were selected and subjected to same well log data as the AI model to estimate water saturation. The estimated water saturation values for AI and other saturation models were then compared with experimental values of testing data and the results showed that AI model was able to predict water saturation with an error of less than 5% while the saturation models did the same with lesser accuracy of error up to 50%. This work clearly shows that computer-based machine learning techniques can determine water saturation with a high precision and the developed correlation works extremely well in prediction mode.","PeriodicalId":10909,"journal":{"name":"Day 2 Tue, October 01, 2019","volume":"308 2 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72950018","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yeek Huey Ho, Nor Baizurah Ahmad Tajuddin, Muhammed Mansor Elharith, H. Dan, Kwang Chian Chiew, Kok Liang Tan, R. Tewari, R. Masoudi
Managing a 47-year brownfield, offshore Sarawak, with thin remaining oil rims has been a great challenge. The dynamic oil rim movement has remained as a key subsurface uncertainty especially with the commencing of redevelopment project. A Reservoir, Well and Facilities Management (RWFM) plan was detailed out to further optimize the development decisions. This paper is a continuation from SPE-174638-MS and outlines the outcome of the RWFM plan and the results’ impact towards the development decisions, such as infill well placement and gas/water injection scheme optimization. Key decisions impact by the RWFM findings are highlighted. One of the RWFM plans is oil rim monitoring through saturation logging to locate the current gas-oil contact (GOC) and oil-water contact (OWC). Cased-hole saturation logs were acquired at the identified observation-wells across the reservoir to map time-lapse oil rim movement and its thickness distribution. Pressure monitoring with regular static pressure gradient surveys (SGS) as well as production data, helped to understand the balance of aquifer strength between the Eastern and Western flanks. Data acquisition opportunity during infill drilling were also fully utilized to collect more solid evidences on oil rim positions, where extensive data acquisition program, including conventional open-hole log, wireline pressure test, formation pressure while drilling (FPWD) and reservoir mapping-while-drilling, were implemented. The timely collection, analysis and assimilation of data helped the team to re-strategize the development / reservoir management plans, through the following major activities: Re-strategizing water and gas injection plan to balance back oil rim between the Eastern and Western flanks, through deferment of drilling water injectors, optimization of water and gas injectors location and completion strategies due to stronger aquifer encroachment from east and south east.Optimizing infill wells drainage points where 2 wells were relocated based on cased-hole logs, as the first well original location was swept and the second well was successfully navigated through the oil rim using reservoir mapping-while-drilling techniques coupled with cased-hole log results. This resulted in securing an oil gain of 4000 BOPD from these 2 wells.Optimizing infill wells location and planning an additional infill well with potential additional oil gain of approximately 2000 BOPD.The understanding of current contact and aquifer strength from the surveillance data assisted in identifying fit-for-purpose technology for the new wells such as the application of viscosity-based autonomous inflow control device which assisted in placing the well closer to GOC due to the observed rapid rising of water table, this will help sustaining the well life. This paper highlights the importance of data integration from geological knowledge, production history, reservoir understanding and monitoring through regular SGS and time-lapse cased-
{"title":"Value Creation from the Reservoir, Well and Facilities Management RWFM Planning in Multi-Stacked Mature Oil Rim Reservoirs, Offshore Sarawak Malaysia","authors":"Yeek Huey Ho, Nor Baizurah Ahmad Tajuddin, Muhammed Mansor Elharith, H. Dan, Kwang Chian Chiew, Kok Liang Tan, R. Tewari, R. Masoudi","doi":"10.2118/196208-ms","DOIUrl":"https://doi.org/10.2118/196208-ms","url":null,"abstract":"\u0000 Managing a 47-year brownfield, offshore Sarawak, with thin remaining oil rims has been a great challenge. The dynamic oil rim movement has remained as a key subsurface uncertainty especially with the commencing of redevelopment project. A Reservoir, Well and Facilities Management (RWFM) plan was detailed out to further optimize the development decisions. This paper is a continuation from SPE-174638-MS and outlines the outcome of the RWFM plan and the results’ impact towards the development decisions, such as infill well placement and gas/water injection scheme optimization. Key decisions impact by the RWFM findings are highlighted.\u0000 One of the RWFM plans is oil rim monitoring through saturation logging to locate the current gas-oil contact (GOC) and oil-water contact (OWC). Cased-hole saturation logs were acquired at the identified observation-wells across the reservoir to map time-lapse oil rim movement and its thickness distribution. Pressure monitoring with regular static pressure gradient surveys (SGS) as well as production data, helped to understand the balance of aquifer strength between the Eastern and Western flanks. Data acquisition opportunity during infill drilling were also fully utilized to collect more solid evidences on oil rim positions, where extensive data acquisition program, including conventional open-hole log, wireline pressure test, formation pressure while drilling (FPWD) and reservoir mapping-while-drilling, were implemented.\u0000 The timely collection, analysis and assimilation of data helped the team to re-strategize the development / reservoir management plans, through the following major activities: Re-strategizing water and gas injection plan to balance back oil rim between the Eastern and Western flanks, through deferment of drilling water injectors, optimization of water and gas injectors location and completion strategies due to stronger aquifer encroachment from east and south east.Optimizing infill wells drainage points where 2 wells were relocated based on cased-hole logs, as the first well original location was swept and the second well was successfully navigated through the oil rim using reservoir mapping-while-drilling techniques coupled with cased-hole log results. This resulted in securing an oil gain of 4000 BOPD from these 2 wells.Optimizing infill wells location and planning an additional infill well with potential additional oil gain of approximately 2000 BOPD.The understanding of current contact and aquifer strength from the surveillance data assisted in identifying fit-for-purpose technology for the new wells such as the application of viscosity-based autonomous inflow control device which assisted in placing the well closer to GOC due to the observed rapid rising of water table, this will help sustaining the well life.\u0000 This paper highlights the importance of data integration from geological knowledge, production history, reservoir understanding and monitoring through regular SGS and time-lapse cased-","PeriodicalId":10909,"journal":{"name":"Day 2 Tue, October 01, 2019","volume":"31 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73186612","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In the absence of well-developed calibrated geologic and simulation models, empirical approaches such as decline curve analysis (DCA) are normally used for production forecasting and reserve estimation. DCA is computationally more efficient compared to simulation models when the active well base exceeds hundreds of wells. However, the underlying assumption for conventional DCA is no change in well operation settings. Moreover, the common approach for production forecasting consists of manual outlier detection and removal, interpretation of missing measurements and data fitting using different models for each well. Therefore, the process of conventional DCA is subjective due to the lack of a standard workflow for preprocessing and data cleansing. The common practice for doing DCA has three main steps: 1. Finding the most representative period in the history of well, 2. Detecting the initial rate (start point) of forecast, 3. Selecting the type of decline and fitting the appropriate model to data points. The solutions to these problems could vary from engineer to engineer and it can be time consuming to analyze all wells manually. To address these issues, we developed a novel workflow based on stochastic methods for detecting various well interventions including change in artificial lift, pump changes and acid treatment, and for forecasting oil production rate more accurately in the presence of uncertainty. The novelty of the proposed ensemble-based approach is forecasting conditioned on various well interventions. Furthermore, the proposed unsupervised stochastic anomaly detection method will detect various well works (or events) in the case of missing records of time and type of events. In this paper, we designed two experiments to test the proposed workflow for oil production rate forecasting and evaluation of acid treatments.
{"title":"A Novel Workflow for Oil Production Forecasting using Ensemble-Based Decline Curve Analysis","authors":"Siavash Hakim Elahi","doi":"10.2118/195916-ms","DOIUrl":"https://doi.org/10.2118/195916-ms","url":null,"abstract":"\u0000 In the absence of well-developed calibrated geologic and simulation models, empirical approaches such as decline curve analysis (DCA) are normally used for production forecasting and reserve estimation. DCA is computationally more efficient compared to simulation models when the active well base exceeds hundreds of wells. However, the underlying assumption for conventional DCA is no change in well operation settings. Moreover, the common approach for production forecasting consists of manual outlier detection and removal, interpretation of missing measurements and data fitting using different models for each well. Therefore, the process of conventional DCA is subjective due to the lack of a standard workflow for preprocessing and data cleansing. The common practice for doing DCA has three main steps: 1. Finding the most representative period in the history of well, 2. Detecting the initial rate (start point) of forecast, 3. Selecting the type of decline and fitting the appropriate model to data points. The solutions to these problems could vary from engineer to engineer and it can be time consuming to analyze all wells manually. To address these issues, we developed a novel workflow based on stochastic methods for detecting various well interventions including change in artificial lift, pump changes and acid treatment, and for forecasting oil production rate more accurately in the presence of uncertainty. The novelty of the proposed ensemble-based approach is forecasting conditioned on various well interventions. Furthermore, the proposed unsupervised stochastic anomaly detection method will detect various well works (or events) in the case of missing records of time and type of events. In this paper, we designed two experiments to test the proposed workflow for oil production rate forecasting and evaluation of acid treatments.","PeriodicalId":10909,"journal":{"name":"Day 2 Tue, October 01, 2019","volume":"199 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72816561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
W. Zaluski, D. Andjelković, Cindy Xu, J. Rivero, M. Faskhoodi, H. Lahmar, Herman Mukisa, Hanatu Kadir, C. Ibelegbu, Warren Pearson, Raouf Ameuri, W. Sawchuk
Enhanced oil recovery (EOR) is an economic way of producing the remaining oil out of previously produced Devonian Pinnacle Reefs in the Nisku Formation within the Bigoray area of Alberta. To maximize the recovery factor of the remaining oil, it was necessary to first characterize the geological structure, matrix reservoir properties, vugular porosity and the natural fracture network of these two carbonate reefs. This characterization model was then used for reservoir simulation history matching and production forecasting further discussed by (Rivero, 2019). With the enhanced resolution of a reprocessed 3D seismic volume, more accurate seismic interpretation was completed to better delineate the internal and external structure of the reefs. The petrophysical analysis and core interpretation showed that the two reefs could be divided into two zones; the bottom zone has low porosity and the upper zone has high porosity that was targeted in previous well completion schemes. These zones were easily picked on well logs and when using Seismic Ant Tracking attributes, were accurately interpreted within the seismic volume. With the framework of the geomodel developed, rock type, porosity, permeability and water saturation were interpolated within the reservoir. Because natural fractures in these carbonate reservoirs are known to be an important part of fluid movement, it was important to characterize the discrete fracture network. In one well, a borehole image successfully quantified the properties of the natural fracture network. The observed fracture density (5 fractures/m) suggested discreate fracture zones throughout the well which was also confirmed with core fracture mapping. As part of the geomodel, a discrete fracture model (DFN) was generated; Seismic Ant Tracking was used to interpolate the fracture intensity within the reservoir. In these Devonian Pinnacle Reefs, and in other reservoirs, before investing in an EOR scheme, it is critical for the operator to understand the geologic structure and the petrophysical characteristics of the reservoir in as much detail as possible. This paper demonstrates how log and seismic data that is up to 40 years old can be converted to modern data types and be used to characterize a reservoir in a way not possible before.
{"title":"An Integrated Approach in Characterization of Triple Porosity Nisku Reefs Alberta: A Quest from Core and Borehole Images to 3D Earth Model","authors":"W. Zaluski, D. Andjelković, Cindy Xu, J. Rivero, M. Faskhoodi, H. Lahmar, Herman Mukisa, Hanatu Kadir, C. Ibelegbu, Warren Pearson, Raouf Ameuri, W. Sawchuk","doi":"10.2118/195882-ms","DOIUrl":"https://doi.org/10.2118/195882-ms","url":null,"abstract":"\u0000 Enhanced oil recovery (EOR) is an economic way of producing the remaining oil out of previously produced Devonian Pinnacle Reefs in the Nisku Formation within the Bigoray area of Alberta. To maximize the recovery factor of the remaining oil, it was necessary to first characterize the geological structure, matrix reservoir properties, vugular porosity and the natural fracture network of these two carbonate reefs. This characterization model was then used for reservoir simulation history matching and production forecasting further discussed by (Rivero, 2019). With the enhanced resolution of a reprocessed 3D seismic volume, more accurate seismic interpretation was completed to better delineate the internal and external structure of the reefs. The petrophysical analysis and core interpretation showed that the two reefs could be divided into two zones; the bottom zone has low porosity and the upper zone has high porosity that was targeted in previous well completion schemes. These zones were easily picked on well logs and when using Seismic Ant Tracking attributes, were accurately interpreted within the seismic volume. With the framework of the geomodel developed, rock type, porosity, permeability and water saturation were interpolated within the reservoir. Because natural fractures in these carbonate reservoirs are known to be an important part of fluid movement, it was important to characterize the discrete fracture network. In one well, a borehole image successfully quantified the properties of the natural fracture network. The observed fracture density (5 fractures/m) suggested discreate fracture zones throughout the well which was also confirmed with core fracture mapping. As part of the geomodel, a discrete fracture model (DFN) was generated; Seismic Ant Tracking was used to interpolate the fracture intensity within the reservoir. In these Devonian Pinnacle Reefs, and in other reservoirs, before investing in an EOR scheme, it is critical for the operator to understand the geologic structure and the petrophysical characteristics of the reservoir in as much detail as possible. This paper demonstrates how log and seismic data that is up to 40 years old can be converted to modern data types and be used to characterize a reservoir in a way not possible before.","PeriodicalId":10909,"journal":{"name":"Day 2 Tue, October 01, 2019","volume":"35 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73413804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Objectives/Scope: In order to maximize the recovery of hydrocarbons from liquids rich shale reservoir systems, the cause and effect relationships between production and the stimulation methods need to be clearly understood. In this study, we integrate a production data regression approach with flow simulation methods to understand the fractured well production behavior and field wide well performance in a liquids rich petroleum system in the Duvernay Basin. Methods, Procedures, Process: Statistical models assume no physical relationship between the model parameters and the response variable, which in this case is produced volumes over a period of time. On the other hand, simulation studies incorporate physical mechanisms of flow to model and predict the production behavior. The simulation models, however, fall short of incorporating all the mechanisms contributing to the production behavior in the complex shale gas reservoir. Thus there is a need for integration of statistical approaches of understanding production behavior along with physics based model and simulation approach. Results, Observations, Conclusions: Multivariate linear regression analysis of the 6 month produced volume and its relationship with parameters such as fracture fluid volumes used, proppant weight placed, and number of stages fractured provides a model with reasonably good correlation. The 6 month produced volumes correlate with large proppant weights, lower fluid placements and greater density of fracture stages. Use of Random Forests machine learning algorithm on the dataset confirms that the total proppant placed, well length completed with fractures have high importance coefficients. In order to examine the well performance using full physical models, fractured well simulations were performed on particular wells using the trilinear model. The trilinear model predictions were compared against other production analyses and the regression model results for consistency. The models showed that in the absence of stress dependent permeability, the production forecast was much higher. Thus, stress dependent permeability appears to be an important factor in the modeling and prediction of production from liquids rich shale reservoirs. Novel/Additive Information: In this study we describe a method to understand the production data from a liquids rich shale reservoir, by integrating multivariate linear regression analysis, machine learning algorithms along with physical model simulations. The results are novel and offer a method to validate either approach to understand cause and effect relationships. This approach may be classified as a new hybrid modeling approach that may potentially be used to optimize stimulation techniques in liquids rich shale reservoirs.
{"title":"Hybrid Methods for Analysis of Fractured Well Production from Liquids Rich Duvernay Shale","authors":"J. Mahadevan, Huanzhen Hu","doi":"10.2118/195798-ms","DOIUrl":"https://doi.org/10.2118/195798-ms","url":null,"abstract":"\u0000 Objectives/Scope: In order to maximize the recovery of hydrocarbons from liquids rich shale reservoir systems, the cause and effect relationships between production and the stimulation methods need to be clearly understood. In this study, we integrate a production data regression approach with flow simulation methods to understand the fractured well production behavior and field wide well performance in a liquids rich petroleum system in the Duvernay Basin.\u0000 Methods, Procedures, Process: Statistical models assume no physical relationship between the model parameters and the response variable, which in this case is produced volumes over a period of time. On the other hand, simulation studies incorporate physical mechanisms of flow to model and predict the production behavior. The simulation models, however, fall short of incorporating all the mechanisms contributing to the production behavior in the complex shale gas reservoir. Thus there is a need for integration of statistical approaches of understanding production behavior along with physics based model and simulation approach.\u0000 Results, Observations, Conclusions: Multivariate linear regression analysis of the 6 month produced volume and its relationship with parameters such as fracture fluid volumes used, proppant weight placed, and number of stages fractured provides a model with reasonably good correlation. The 6 month produced volumes correlate with large proppant weights, lower fluid placements and greater density of fracture stages. Use of Random Forests machine learning algorithm on the dataset confirms that the total proppant placed, well length completed with fractures have high importance coefficients. In order to examine the well performance using full physical models, fractured well simulations were performed on particular wells using the trilinear model. The trilinear model predictions were compared against other production analyses and the regression model results for consistency. The models showed that in the absence of stress dependent permeability, the production forecast was much higher. Thus, stress dependent permeability appears to be an important factor in the modeling and prediction of production from liquids rich shale reservoirs.\u0000 Novel/Additive Information: In this study we describe a method to understand the production data from a liquids rich shale reservoir, by integrating multivariate linear regression analysis, machine learning algorithms along with physical model simulations. The results are novel and offer a method to validate either approach to understand cause and effect relationships. This approach may be classified as a new hybrid modeling approach that may potentially be used to optimize stimulation techniques in liquids rich shale reservoirs.","PeriodicalId":10909,"journal":{"name":"Day 2 Tue, October 01, 2019","volume":"77 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2019-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74526908","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：