Using Neural Network Predictive Control for Riser-Slugging Suppression

Eslamloueyan, Reza; Hosseinzadeh, Elham

doi:10.2202/1934-2659.1315

Cited by 5 publications

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several controllers have been designed for slugging suppression including PI [2], cascaded PID [3], Model Predictive Control (MPC) [4], neural networks [5], and gain-scheduling Internal Model Control (IMC) [6]. These controllers generally attempt to control the pressure at the base of the riser.…”

Section: Introductionmentioning

confidence: 99%

Post-Installed Fiber Optic Pressure Sensors on Subsea Production Risers for Severe Slugging Control

Eaton

Safdarnejad

Hedengren

et al. 2015

Volume 5B: Pipeline and Riser Technology

View full text Add to dashboard Cite

Fiber optic sensors have gained increasing use in monitoring offshore structures. The sensors have successfully monitored flowlines, umbilicals, wells, Tension Leg Platform (TLP) tendons, production and drilling risers, and mooring lines. Fiber optic sensors are capable of monitoring strain, temperature, pressure, and vibration. While the success of fiber optic monitoring has been clearly demonstrated, the sensors are now under consideration for automation applications. This paper details the plausibility of using pressure measurements from post-installed fiber Bragg grating (FBG) sensors with Model Predictive Control (MPC) to suppress severe slugging in subsea risers. Prior control schemes demonstrate that slugging is mitigated using a topside choke valve. The most effective methods use a pressure measurement immediately upstream of the touchdown zone of the riser; however, the majority of production risers do not have pressure sensing at that location. With advances in subsea clamp design and bonding it is now possible to install a non-penetrating FBG sensor to monitor pressure near the touchdown zone without shutting down production. Stabilizing the two phase flow both reduces vibration-induced fatigue and has the potential to allow increased throughput with relaxed topside processing constraints. MPC predicts and adjusts for disturbances to avoid pressure and flow instability. The performance of the controller is influenced by sensor location, choke valve response time, and riser geometry. This study demonstrates that severe riser slugging is effectively controlled with MPC and a post-installed, non-penetrating FBG sensor.

show abstract

Section: Introductionmentioning

confidence: 99%

Post-Installed Fiber Optic Pressure Sensors on Subsea Production Risers for Severe Slugging Control

Eaton

Safdarnejad

Hedengren

et al. 2015

Volume 5B: Pipeline and Riser Technology

View full text Add to dashboard Cite

show abstract

Automatic recognition of oil reservoir models from well testing data by using multi-layer perceptron networks

Vaferi¹,

Eslamloueyan²,

Ayatollahi³

2011

Journal of Petroleum Science and Engineering

112

View full text Add to dashboard Cite

Production Optimization in Oil and Gas Wells: A Gated Recurrent Unit Approach to Bottom Hole Flowing Pressure Prediction

Abdullahi,

Ezeh

2024

SPE Nigeria Annual International Conference and Exhibition

View full text Add to dashboard Cite

Production Optimization continues to be a priority in the energy space. Several means have been investigated, particularly machine learning in recent times. One of the aspects machine learning has aided is in the area of future bottomhole flowing pressure prediction (BHFP). It lets asset managers know potential problems and make proactive decisions to facilitate optimal production. This study explores the application of Gated Recurrent Units (GRUs) in oil and gas wells to estimate future BHFP by using optimal features. To properly evaluate the BHFP predictive capability of the GRU model, the model is compared with the use of Long Short-Term Memory (LSTM) networks since LSTM also keeps long-term memory, which aids time series prediction capabilities. Using historical well data, GRU and LSTM models were trained with an optimal number of features, selected by their importance from Recursive Feature Elimination. The results showed the GRU models performed better than LSTM models, averaging a 30% approximate decrease across error metrics. With the GRU forecasts enhanced by optimized features, production managers can make prompt, well-informed decisions. This, in turn, makes it easier to control reservoirs precisely, make proactive modifications to well operations, and cut down on expensive downtime.

show abstract

Using Neural Network Predictive Control for Riser-Slugging Suppression

Cited by 5 publications

References 12 publications

Post-Installed Fiber Optic Pressure Sensors on Subsea Production Risers for Severe Slugging Control

Post-Installed Fiber Optic Pressure Sensors on Subsea Production Risers for Severe Slugging Control

Automatic recognition of oil reservoir models from well testing data by using multi-layer perceptron networks

Production Optimization in Oil and Gas Wells: A Gated Recurrent Unit Approach to Bottom Hole Flowing Pressure Prediction

Contact Info

Product

Resources

About