In the offshore oil and gas industry, treating produced water, which contains contaminants extracted alongside oil and gas from reservoirs, before it can be safely disposed of or reused is essential. Integrating hydrocyclone systems with three-phase separators significantly improves the efficiency of separating the oil, water, and gas components, thereby ensuring compliance with environmental regulations and reducing operational costs. Consequently, the developed of an optimally coordinated control solution for these integrated systems offers benefits such as enhanced efficiency, cost reduction, regulatory compliance, and improved sustainability. This paper focuses on a comparative study of three control design methodologies: an optimized–PI controller, a model reference adaptive controller (MRAC), and a newly developed Immersion and Invariance (I&I) adaptive controller for offshore produced water treatment. These control strategies are evaluated for their effectiveness in regulating key parameters, such as the pressure drop ratios (PDR) and the water level (L) in the presence of parametric uncertainty and external disturbance. The superior performance of the proposed I&I controller is demonstrated via simulations and analysis.

As the hydrocarbon discharge regulation is becoming more stringent, and emerging sensing and digitization technologies are applied in offshore oil and gas (OG) production, it is beneficial to upgrade the existing control systems by integrating these new measurements/information into the control loops. To achieve this goal, some proper control-oriented models need to be developed. Hydrocyclone-based de-oiling systems are commonly used in the produced water treatment after the three-phase separators in offshore OG production. Control-oriented modeling of these systems has proven a challenging task. In this paper, a system identification approach is taken and polynomial-type Hammerstein-Wiener (HW) models of the separation efficiency are identified for a de-oiling hydrocyclone system using the newly installed online oil-in-water (OiW) measurement. An exhaustive search is used to determine the order of the polynomials and transfer functions. The best model found was a Hammerstein model. The model obtained provides the opportunity to extend the existing de-oiling control framework to include the OiW measurement into feedback loops.

With the objective of pursuing eco-friendly and environmentally sustainable extraction in offshore oil and gas production, the Produced Water (PW) treatment is inescapable whether for reinjecting the PW back to the reservoir with the purpose of enhancing oil recovery or discharging it into the ocean. The de-oiling technology based on the hydrocyclones system connected to the three-phase separator is generally employed for the PW treatment. In order to boost the effectiveness of the interconnected de-oiling systems, it is crucial to introduce an advanced coordinated controller for the separators and hydrocyclone systems with regard to the uncertainties and unknown disruptions that occur during dynamic operations. This research suggests a robust Multivariate Model Reference Adaptive Control (MRAC) approach, in order to regulate the Pressure-Drop-Ratio (PDR) of the hydrocyclone and the water level in the separator simultaneously. The suggested control technique is expressed within a framework of state feedback output tracking MRAC by taking adaptive disturbance rejection into account. Furthermore, a control parametrizing obtained from a factorization of the High-Frequency Gain Matrix (HFGM) in the shape of the product of three matrices, i.e. LDS decomposition, is employed in the designing procedure to relax restrictive conditions about the HFGM. Finally, the proposed robust MRAC method is tested on a lab-scaled PW application, and the simulation outcomes explicitly establish the effectiveness and also its remarkable performance of the suggested control strategy.

By deploying the online Oil-in-Water (OiW) sensors in a de-oiling hydrocyclone system used for produced water treatment processes in offshore oil & gas production, this work investigated modelling of the complicated separation dynamics inside the hydrocylone system using the Long-Short-Term Memory Neural Network (LSTM-NN). The purpose of this modelling is to predict the hydrocyclone's transient de-oiling efficiency in a high level of accuracy. Thereby the hydrocyclone system can be optimally controlled subject to different operating conditions. The acquisition and analysis of the data obtained from a lab-scaled pilot plant is introduced. Two types of LSTM-NN configurations are proposed, and the hyper-parameter tuning as well as training and validation results, are discussed in details. The results exhibit that the relative concentration of OiW, which correlated with the de-oiling efficiency, can be predicted in a quite accurate level using two types of measurements, i.e., the opening degrees of cyclone's underflow and overflow control valves, both the hydrocyclone's inlet/water-outlet OiW concentration measurements. The best model can achieve a normalized RMSE 83,62% accuracy in the validation test. One of our next step is to cooperate the LSTM - NN model into the model predictive control framework to design some optimal control solution for de-oiling hydrocyclone systems.

Three different control solutions, named PID, H∞and MPC, are proposed and compared for a de-oiling water treatment process used in offshore oil and gas production. All control solutions are designed using the plant-wide control strategy, i.e., the considered process consists of the upstream three-phase separator(s) and the downstream de-oiling hydorcyclone(s), such that their dynamic coupling is systematically coped with. The system performances are evaluated according to the conventional control criteria as well as the emerging online Oil-in-Water (OiW) measurement. The experimental results based on a lab-scaled pilot plant illustrate the huge potential and benefit of advanced control solutions to significantly improve the effluent water quality without sacrificing the system’s throughput, particularly when the process is subject to large disturbances (e.g., slugging inflow).

This paper establishes an implementation framework, as a proof of concept, of how to reduce the uncertainty of deploying advanced control offshore. The majority of process research tends to consider improvements in performance, but with less emphasis on how to realize implementation. For control methods to be successfully applied offshore, the methods must be sufficiently simple, trustworthy, and transparent. This is mainly due to the severe consequence of incidences offshore. As it is ultimately the operators that decide which control methods are toggled on/off, the operators need to be aware of the control's behavior. The focus of this paper is not process performance, nor control theory, but rather how to convey the status, state, and action of the controllers to the offshore operators. A design approach is given for displaying and explaining the control for the operators. The is based on uniting the fast prototyping capability of Simulink Real-Time with the graphical capabilities of a Human Machine Interface system. As a case study, experiments are carried out to compare Model Predictive Control to conventional Proportional Integral Derivative control on a scaled offshore pilot-plant, which can emulate different separation processes at the topside of offshore oil & gas installations. The results show that the established connection makes it possible to investigate and compare control systems real-time, which data should be available to an operator and how to represent it.