-
Complementarity-constrained predictive control for efficient gas-balanced hybrid power systems
Authors:
Kiet Tuan Hoang,
Brage Rugstad Knudsen,
Lars Struen Imsland
Abstract:
Controlling gas turbines (GTs) efficiently is vital as GTs are used to balance power in onshore/offshore hybrid power systems with variable renewable energy and energy storage. However, predictive control of GTs is non-trivial when formulated as a dynamic optimisation problem due to the semi-continuous operating regions of GTs, which must be included to ensure complete combustion and high fuel eff…
▽ More
Controlling gas turbines (GTs) efficiently is vital as GTs are used to balance power in onshore/offshore hybrid power systems with variable renewable energy and energy storage. However, predictive control of GTs is non-trivial when formulated as a dynamic optimisation problem due to the semi-continuous operating regions of GTs, which must be included to ensure complete combustion and high fuel efficiency. This paper studies two approaches for handling the semi-continuous operating regions of GTs in hybrid power systems through predictive control, dynamic optimisation, and complementarity constraints. The proposed solutions are qualitatively investigated and compared with baseline controllers in a case study involving GTs, offshore wind, and batteries. While one of the baseline controllers considers fuel efficiency, it employs a continuous formulation, which results in lower efficiency than the two proposed approaches as it does not account for the semi-continuous operating regions of each GT.
△ Less
Submitted 10 April, 2024;
originally announced April 2024.
-
Physics-Informed Neural Networks with Skip Connections for Modeling and Control of Gas-Lifted Oil Wells
Authors:
Jonas Ekeland Kittelsen,
Eric Aislan Antonelo,
Eduardo Camponogara,
Lars Struen Imsland
Abstract:
Neural networks, while powerful, often lack interpretability. Physics-Informed Neural Networks (PINNs) address this limitation by incorporating physics laws into the loss function, making them applicable to solving Ordinary Differential Equations (ODEs) and Partial Differential Equations (PDEs). The recently introduced PINC framework extends PINNs to control applications, allowing for open-ended l…
▽ More
Neural networks, while powerful, often lack interpretability. Physics-Informed Neural Networks (PINNs) address this limitation by incorporating physics laws into the loss function, making them applicable to solving Ordinary Differential Equations (ODEs) and Partial Differential Equations (PDEs). The recently introduced PINC framework extends PINNs to control applications, allowing for open-ended long-range prediction and control of dynamic systems. In this work, we enhance PINC for modeling highly nonlinear systems such as gas-lifted oil wells. By introducing skip connections in the PINC network and refining certain terms in the ODE, we achieve more accurate gradients during training, resulting in an effective modeling process for the oil well system. Our proposed improved PINC demonstrates superior performance, reducing the validation prediction error by an average of 67% in the oil well application and significantly enhancing gradient flow through the network layers, increasing its magnitude by four orders of magnitude compared to the original PINC. Furthermore, experiments showcase the efficacy of Model Predictive Control (MPC) in regulating the bottom-hole pressure of the oil well using the improved PINC model, even in the presence of noisy measurements.
△ Less
Submitted 4 March, 2024;
originally announced March 2024.
-
Multi-unit soft sensing permits few-shot learning
Authors:
Bjarne Grimstad,
Kristian Løvland,
Lars S. Imsland
Abstract:
Recent literature has explored various ways to improve soft sensors by utilizing learning algorithms with transferability. A performance gain is generally attained when knowledge is transferred among strongly related soft sensor learning tasks. A particularly relevant case for transferability is when develo** soft sensors of the same type for similar, but physically different processes or units.…
▽ More
Recent literature has explored various ways to improve soft sensors by utilizing learning algorithms with transferability. A performance gain is generally attained when knowledge is transferred among strongly related soft sensor learning tasks. A particularly relevant case for transferability is when develo** soft sensors of the same type for similar, but physically different processes or units. Then, the data from each unit presents a soft sensor learning task, and it is reasonable to expect strongly related tasks. Applying methods that exploit transferability in this setting leads to what we call multi-unit soft sensing.
This paper formulates multi-unit soft sensing as a probabilistic, hierarchical model, which we implement using a deep neural network. The learning capabilities of the model are studied empirically on a large-scale industrial case by develo** virtual flow meters (a type of soft sensor) for 80 petroleum wells. We investigate how the model generalizes with the number of wells/units. Interestingly, we demonstrate that multi-unit models learned from data from many wells, permit few-shot learning of virtual flow meters for new wells. Surprisingly, regarding the difficulty of the tasks, few-shot learning on 1-3 data points often leads to high performance on new wells.
△ Less
Submitted 13 May, 2024; v1 submitted 27 September, 2023;
originally announced September 2023.
-
Adjustment formulas for learning causal steady-state models from closed-loop operational data
Authors:
Kristian Løvland,
Bjarne Grimstad,
Lars Struen Imsland
Abstract:
Steady-state models which have been learned from historical operational data may be unfit for model-based optimization unless correlations in the training data which are introduced by control are accounted for. Using recent results from work on structural dynamical causal models, we derive a formula for adjusting for this control confounding, enabling the estimation of a causal steady-state model…
▽ More
Steady-state models which have been learned from historical operational data may be unfit for model-based optimization unless correlations in the training data which are introduced by control are accounted for. Using recent results from work on structural dynamical causal models, we derive a formula for adjusting for this control confounding, enabling the estimation of a causal steady-state model from closed-loop steady-state data. The formula assumes that the available data have been gathered under some fixed control law. It works by estimating and taking into account the disturbance which the controller is trying to counteract, and enables learning from data gathered under both feedforward and feedback control.
△ Less
Submitted 10 November, 2022;
originally announced November 2022.
-
Bayesian Neural Networks for Virtual Flow Metering: An Empirical Study
Authors:
Bjarne Grimstad,
Mathilde Hotvedt,
Anders T. Sandnes,
Odd Kolbjørnsen,
Lars S. Imsland
Abstract:
Recent works have presented promising results from the application of machine learning (ML) to the modeling of flow rates in oil and gas wells. Encouraging results and advantageous properties of ML models, such as computationally cheap evaluation and ease of calibration to new data, have sparked optimism for the development of data-driven virtual flow meters (VFMs). Data-driven VFMs are developed…
▽ More
Recent works have presented promising results from the application of machine learning (ML) to the modeling of flow rates in oil and gas wells. Encouraging results and advantageous properties of ML models, such as computationally cheap evaluation and ease of calibration to new data, have sparked optimism for the development of data-driven virtual flow meters (VFMs). Data-driven VFMs are developed in the small data regime, where it is important to question the uncertainty and robustness of models. The modeling of uncertainty may help to build trust in models, which is a prerequisite for industrial applications. The contribution of this paper is the introduction of a probabilistic VFM based on Bayesian neural networks. Uncertainty in the model and measurements is described, and the paper shows how to perform approximate Bayesian inference using variational inference. The method is studied by modeling on a large and heterogeneous dataset, consisting of 60 wells across five different oil and gas assets. The predictive performance is analyzed on historical and future test data, where an average error of 4-6% and 8-13% is achieved for the 50% best performing models, respectively. Variational inference appears to provide more robust predictions than the reference approach on future data. Prediction performance and uncertainty calibration is explored in detail and discussed in light of four data challenges. The findings motivate the development of alternative strategies to improve the robustness of data-driven VFMs.
△ Less
Submitted 13 June, 2021; v1 submitted 2 February, 2021;
originally announced February 2021.
