Field Surveillance and AI based Steam Allocation Optimization Workflow for Mature Brownfield Steam Floods

Kumar, Arvind; Novlesky, Alex; Bityutsky, Erykah; Koci, Paul; Wightman, Jeff

doi:10.2118/193700-ms

Cited by 9 publications

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A simulation-augmented data-driven approach for cyclic steam development is proposed to include field data analysis and integration and mechanistic modeling of the field in interest, thus generating a machine-learning model to evaluate various scenarios, optimize injection job parameters, and assess uncertainties [36]. In addition, machine-learning-assisted field applications, e.g., field surveillance for steam floods, predicting heavy-oil combustion kinetics, and optimization of steam-injection plan in a real field, also verify its ability in multiple petroleum-related projects [37][38][39].…”

Section: Introductionmentioning

confidence: 93%

Machine-Learning-Assisted Identification of Steam Channeling after Cyclic Steam Stimulation in Heavy-Oil Reservoirs

Liu

Jiao

et al. 2023

Geofluids

View full text Add to dashboard Cite

Cyclic steam stimulation (CSS) is one efficient technology for enhancing heavy-oil recovery. However, after multiple cycles, steam channeling severely limits the thermal recovery because high-temperature steam preferentially breaks through to the producers. To solve the issues of steam breakthrough, it is essentially important and necessary to recognize steam channeling. In this work, a machine-learning-assisted identification model, based on a random-forest ensemble algorithm, is developed to predict the occurrence of steam channeling during steam huff-and-puff processes. The set of feature attributes is constructed based on the permeability ratio, steam quality, and steam-injection speed, which provides the reference for the construction of the training-sample set, steam-channeling reconstruction set, and prediction set. Based on the realistic data, the Pearson correlation coefficient is implemented to confirm the linear correlation among different characteristics; thus, the dimension reduction of the characteristic parameters is achieved. The random oversampling method is adopted to treat the unbalanced training-sample set. Our results show that this model can accurately describe the current state of steam channeling and predict steam propagation in the following cycles.

show abstract

Section: Introductionmentioning

confidence: 93%

Machine-Learning-Assisted Identification of Steam Channeling after Cyclic Steam Stimulation in Heavy-Oil Reservoirs

Liu

Jiao

et al. 2023

Geofluids

View full text Add to dashboard Cite

show abstract

Enhancing Reservoir Management Quality and Efficiency of Thermal Assets with Data-Driven Models

Kim

2019

SPE Western Regional Meeting

View full text Add to dashboard Cite

Temperature monitoring is the most important surveillance in thermal assets, but temperature logging is limited in frequencies and locations. In addition, it is extremely difficult to review all the measured temperature and injection data manually since there are 10,000+ wells in Kern River field. To overcome the limitations, data-driven reservoir temperature models are presented that are built using past temperature logs and steam injection rates of the Kern River field, California. Based on the physics and geologic understanding the reservoir, adequate input features were selected and queried. Data cleanup was conducted to remove erroneous data or fix data errors using statistical tools such as multivariate Gaussian distribution. Voronoi diagram based dynamic injector selection algorithm (DISA) was developed to correctly capture the injectors which impact on temperature changes of a temperature observation well. Based on geologic characteristics of the Kern River, reservoir was divided into two sub-reservoirs, North-East and South-West. Two full field models were developed for predicting maximum and mean temperatures of a heated zone with multi-layer perceptron for both sub-reservoirs using about 120,000 data points from over 25,000 temperature curves measured at 700+ temperature observation wells. To estimate proper model update frequencies and verify the process, three yearly models (models 2015, 2016, and 2017) were built and validated by using one-year future temperature predictions in 2016, 2017, and 2018. For instance, model 2015 was trained with data until the end of 2015 and validated against 2016 data. Maximum temperature prediction r2 of 2017 South-West and North-East models were 0.96 and 0.98, respectively. Model 2017 has been deployed for alerting exception cases automatically and flagging abnormal temperature measurements. Also, the models improve the quality of heat injection design by providing temperature predictions based on planned heat injection rates. This novel automated workflow with data-driven models enhances reservoir management efficiency by reducing engineers’ unproductive time such as data manipulation and allowing them to focus on value-added works like analysis and optimization.

show abstract

Real-Time Steam Allocation Workflow Using Machine Learning for Digital Heavy Oil Reservoirs

Sibaweihi

Patel

Guevara

et al. 2019

SPE Western Regional Meeting

View full text Add to dashboard Cite

Thermal oil recovery processes are widely used to extract bitumen and heavy oil. Traditionally, a predetermined amount of steam is allocated to various injector wells using reservoir model based open-loop optimization. This practice can face a number of constraints including interruptions in well operations and/or surface facilities. Given that steam supply costs are a significant contributor to the overall production cost of heavy oil, dynamic and intelligent allocation of steam to various wells in the oilfield deserves further attention. In this study, we propose a proactive steam allocation workflow that can learn the effect of steam injection pattern on heavy oil recovery by using machine learning. We employ data analytic predictive models for the short-term forecast of the key performance indicators (KPIs). Model parameters are updated continuously by using a moving horizon approach that considers selected prior data including real-time measurements. An objective function containing predicted KPIs is maximized by manipulating the amount of steam allocated to various injectors in the oilfield. The workflow is repeated on a daily basis for continuous optimum steam allocation. A case study is performed by using a 3D reservoir model that represents a segment of the steam-assisted gravity drainage (SAGD) operation. For each well, the polynomial model is identified in the time-domain to forecast KPIs. The effectiveness of the proposed method is evident from the results as NPV is increased by almost 25% – 50% compared to the base case with a constant steam injection pattern in all cases studied. Due to the efficient use of available steam, the steam-to-oil ratio is reduced significantly. An adaptive and flexible steam supply is also honored by the proposed workflow, ensuring maximum efficiency of the oil recovery process. Practical implications of the proposed intelligent steam allocation workflow will be consequential in improving the operational efficiency of the digital heavy oil assets, thereby increasing profits and reducing the carbon footprint.

show abstract

Field Surveillance and AI based Steam Allocation Optimization Workflow for Mature Brownfield Steam Floods

Cited by 9 publications

References 11 publications

Machine-Learning-Assisted Identification of Steam Channeling after Cyclic Steam Stimulation in Heavy-Oil Reservoirs

Machine-Learning-Assisted Identification of Steam Channeling after Cyclic Steam Stimulation in Heavy-Oil Reservoirs

Enhancing Reservoir Management Quality and Efficiency of Thermal Assets with Data-Driven Models

Real-Time Steam Allocation Workflow Using Machine Learning for Digital Heavy Oil Reservoirs

Contact Info

Product

Resources

About