This work presents the development of an advanced control strategy using Model Predictive Control (MPC) for controlling the gas compression system of an offshore platform. It includes details about the complete phenomenological model of the system and of the software infrastructure developed to support the system implementation in real conditions. The proposed control structure has two main goals: (i) avoid unwanted regions of operation; and (ii) increase stability margins and availability. These goals are achieved by using a zone-control MPC and by adequately interacting with the regulatory control level. Although the proposed structure is general, this work exemplifies its application in a particular compression unit of a real offshore platform. Simulation results are presented in two different scenarios, one to test how the controller rejects a gas-load varying disturbance and another to analyze how the controller copes with an abnormal situation, losing real-time data of process variables or manipulated variables during operation. The good performance obtained in these two cases confirm the benefits provided by the proposed MPC strategy to the operation of the gas compression unit.

This paper presents the use of a radial basis function artificial neural network to estimate sensor readings exploring the analytical redundancy via auto-association. However, in order to guarantee optimal performance of the network, the training and optimization processes have been modified. In the conventional training algorithm, even if a stop criterion, such as summed squared error, is reached, one or more of the individual performance metrics, including: i) accuracy; ii) robustness; iii) spillover and iv) filtering of the neural network, may not be satisfactory while validating sensor measurements. Essentially, the proposed modification in the training algorithm is based on seeking to ensure that one or more of the metrics are met. This paper describes the proposed algorithm including all of its mathematical foundation. Afterward, a data set of a water injection pump for an oil and gas processing unit was used to train the RBF network using the conventional and the modified algorithm, and the performance of each was evaluated. Furthermore, the AAKR model is applied to the same dataset as a quality reference parameter. Finally, a comparison analysis of the developed models is presented for each of the performance metrics, as well as for overall effectiveness, demonstrating that the main advantage of the proposed approach is to obtain the estimation results equivalent or superior to the AAKR with shorter runtime and the disadvantage of having higher complexity during the model training.

The natural gas produced in primary separation is passed through a compression system to be pressurized and conditioned before being sent to its final destination. The operation of that system needs to be safe and efficient to avoid equipment damage and reduce energy consumption. The stable and secure operation of the equipment in a compression system is provided, many times, by a classical regulatory control layer. In this work, we present a Model Predictive Control (MPC) strategy to provide setpoints for the regulatory control layer of a gas compression system, aiming to avoid excessive energy consumption, decrease the plant variability, and guarantee a stable and safe operation against load disturbances. The proposed method is tested in a digital twin of a typical industrial unit using a Dead-Time Compensator Generalized Predictive Controller (DTC-GPC). Some disturbances in the feed flow rate of gas were considered as case studies. The controller responded satisfactorily to these disturbances keeping the plant operation stable and returning the controlled variables in the desired operating range after a short time.

The flow rate values reported in real time, for an oil and gas production unit, refer to the total amount produced by that unit. On the other hand, data referring to the flow rate of each producing well in real time are usually not available, due to the difficulty in implementing flow measurement devices. These individual flows are determined by production tests that are generally performed every two months and are important for production planning and optimization. Therefore, there is a need to generate models capable of predicting the flow of each well in the period between tests, in order to identify possible issues during the production. In this context, we propose in this paper a method to predict the liquid and gas flow rates of each well as function of measured variables available in plant data collected in real time. Choke valve specifications and fluid properties are also required as model input data. Since there is no flow measurement device, it is not possible to directly validate the model results against plant data. In this case, the model validation is performed using the total liquid and gas flow rates presented in the Daily Operation Report (DOR) and those provided by fiscal meters. Relative errors below 3.5% were observed, showing good agreement between the calculated flow rates and the provided by fiscal meters. The proposed method was implemented in the computational environment EMSO, which presents as advantages the low simulation time and robustness over a wide range of input data. These advantages become important as the model is used for monitoring platforms in real time. The range of applicability of this model can be attributed to the Gas-Oil Rate (GOR). In this work, the lowest and highest values of GOR considering all the wells are 207.58 and 393.97, respectively.

This work presents the use of radial basis function artificial neural network to estimate the sensors measurements, exploring the analytical redundancy existent among different sensors in a process. However, in order to guarantee good performance of the network the training and optimization process was modified. In the conventional training algorithm, although the stop criteria, such as summed squared error, is reached, one or more of the individual performance metrics of the neural network may not be satisfactory. The performance metrics considered are Accuracy (training error), Sensitivity matrix (sensors propagated error to the estimations) and Filtering matrix (sensor propagated noise to the estimations). The paper describes the proposed method including all the mathematical foundation. A dataset of a petroleum refinery is used to train a RBF (Radial Basis Function) network using the conventional and the modified method and the performance of both will be evaluated. Furthermore, AAKR (Auto-Associative Kernel Regression) model is used to the same dataset. Finally, a comparison study of the developed models will be done for each of the performance metrics, as well as for the overall effectiveness in order to demonstrate the superiority of the proposed approach.

This work presents a modelling methodology for sensors and equipment condition monitoring developed during a research project to enhance dependability of pre-salt petroleum extraction platforms. The methodology aims to improve the capability of the auto-associative models applied for sensors monitoring in the last decades in nuclear power plants, chemical industry, refineries, gas transport and processing plants. However, actual operation problems or fault in equipment also may lead to false measurement error detection. This problem observed in the previous applications motivated the development of the improved method able to detect measurement errors and fault conditions in the process or equipment. This improvement has been obtained adding data (real or simulated) of the different conditions of operation, including the fault conditions (undesired data in the previous methodology). Therefore, the models become able to make accurate sensor estimation, even under fault conditions in the monitored process, and they also give a proper fault diagnoses about the measurement instruments and the process reducing false alarms compared to the traditional approaches. Also, some modelling challenges were observed during the development such as optimization of parameters, memory size and computing complexity. The methodology is demonstrated using simulated a process of a Petroleum Platform application. The achieved results showed a possible methodology to improve or replace the traditional approaches in the past application.

Currently, there is a trend in reduction of the number of industrial plant operators. The challenges are mainly during emergency situations: how to support operator time management without increasing operational risks? SDA focuses on this area and aims to increase operator situational awareness (ability to perceive, understand and predict the future behavior of a process) through new technological paradigms, such as Expert System and Ecological Human Machine Interface (HMI) in order to provide operational support, maintenance and optimization of refining, exploration and system of production of oil and gas plants. In SDA, the most critical alerts are shown by priority, along with decision trees, trend charts and variable comparison charts. SDA aims to assist control room operators in solving a critical problem in the oil industry, that is the loss of safety function, associated with alarms, during alarm flood. The SDA results of the SDA are presented through its implementation in Sulfur Recovery Units—URE, in the state of Rio de Janeiro, in Brazil.

As an offshore oil well ages, it is common for the production system to face multiphase flow problems such as limit cycles. This phenomenon, known as slugging in the jargon of the oil industry, causes oscillations in the well’s flowrate and pressure. Its main effects are reducing production and increasing the risk of operational discontinuity due to shut down. In this paper, an advanced control process (APC) strategy is presented to deal with the slugging problem in oil wells. The strategy uses a two-layer coupled control structure: a regulatory via a PID control, and a supervisory via a model-based predictive control (MPC). The structure proposed was applied to a real ultra-deepwater well in Petrobras that was partially restricted by the choke valve to avoid the propagation of oscillatory behavior to the production system. As a result, the well has achieved a 10% oil production increase while maintaining the flow free of severe slugging, which meant an increment of about 240 barrels a day for that specific well.

This paper presents a real-world application of a mixed-integer receding horizon control in an onshore oil gathering network. The objective is to stabilize the operation of the gathering network by coordinating the automatic switching of pumps to avoid abrupt flow variations at the central station, minimizing pump switching and, at the same time, maintaining the level of fluid in the satellite stations under control. The applied technological solution easily integrates different systems, optimization tools and heuristic rules, allowing in-house development of complex applications.

Goal: Industry 4.0 enables the design of new models for process monitoring in which sensors, analyzers, and controls are positioned at different points in the process. The goal of this work is to present the modeling and control of a three-phase production separator with hydrocyclones to treat produced water on an oil platform of mature fields. Design / Methodology / Approach: The Methodology or approach used was to develop a model for the primary separator that allows its operation by means of a controller (fuzzy and PI) to manipulate the flow of discarded water, acting indirectly in the oil-in-water measure. Results: The results showed the consistency of the model for open loop simulations and the effectiveness of the controllers to comply with the discarding requirements for the closed-loop simulations. Limitations of the investigation: The limitation of the model and developed controllers is that they are only applied to platforms where water production exceeds the separators discharge capacity and the exceeding water can be offloaded to another equipment or to another platform. Practical implications: The main practical implications of this study are to maximize the flow of discarded water on mature field platforms, which produces elevated amounts of water, to conform the total oil and grease to the local law regulations. Additionally, it also increases oil production, with a higher limit of water production. Originality / Value: Compared to the previous authors, where the models of discarded flow is a function of the water-oil interface, this work developed a model that allowed the flow of the discarded water to be a function of its quality.

This work presents the use of radial basis function artificial neural network to estimate the sensors readings, exploring the analytical redundancy via auto association. However, in order to guarantee good performance of the network the training and optimization process was modified. In the conventional training algorithm, although the stop criteria, such as summed squared error, is reached, one or more of the individual performance metrics, including: i) accuracy; ii) robustness; iii) spillover and iv) filtering matrix of the neural network may not be satisfactory. The paper describes the proposed algorithm including all the mathematical foundation. A dataset of a petroleum refinery is used to train a RBF network using the conventional and the modified algorithm and the performance of both will be evaluated. Furthermore, AAKR model is used to the same dataset. Finally, a comparison study of the developed models will be done for each of the performance metrics, as well as for the overall effectiveness in order to demonstrate the superiority of the proposed approach.

Este trabalho trata da determinacao da producao de liquido em cada poco submarino de uma plataforma de producao de petroleo do tipo FPSO (Floating, Production, Storage and Offloading). A utilizacao de medidores de fluxo multifasicos e uma solucao cara e indisponivel na maioria das unidades produtoras e as medidas de variaveis de fundo (pressao e temperatura) possuem baixa confiabilidade. Alem disso, os sensores submersos, em geral, deixar de operar apos alguns anos. Assim, o objetivo desta pesquisa e desenvolver um algoritmo para estimar a vazao de producao em pocos de petroleo, com base somente nas medidas de variaveis de superficie (pressao, temperatura e posicao de valvula). O metodo proposto consiste em

Oil production employing gas lift techniques enable the production of no natural flow wells and supply the energy lost in the reservoir caused by the field depletion, keeping the production in brown fields feasible. The multiphase flow conditions and the long pipes used to transport the fluids from the reservoir to the surface facilities, especially in deep and ultra-deepwater cases, may create unstable flow situations. Several publications in process control have discussed this problem since the 1980s, but the potential multivariable actions on the choke valve and gas lift flow have not been explored so far. In this paper the operating oil production system is treated through a nonlinear predictive control strategy. The strategy evaluation in a rigorous model (OLGA) shows the association between predictive capability and the integrated actuation in the manipulated variables results in an oil production increase and a partial or entire suppression of the instabilities in the multiphase flow. Furthermore, the rate of acting required on the valves is lower in the multivariable approach, allowing the use of slow choke valves as a final control element.

This work describes a simplified dynamic model for control and real time applications in offshore deepwater and ultra-deepwater petroleum production systems. Literature about simplified dynamical models, capable of cover the global architecture of an offshore multiphase production system, is scarce. Hence, the proposed model integrates and adapts partial models available in the literature in order to generate a single model of the whole system. The model, designed to represent slugs generated by the casing heading and terrain/riser concomitantly, was evaluated by comparison with a traditional commercial simulator and was also implemented in two actual production systems. As a result, the model showed the capability of capturing complex dynamical behaviors, such as limit cycles, demonstrated to be numerically more stable than similar models in literature, fast enough to be used in real time applications and proved to be adherent to the commercial simulator and actual operating data from Petrobras production systems.

Abstract It is difficult to control and to manage wells’ start-up in offshore platforms. In order to solve this problem an intelligent system can play an important role, since available qualitative operator and design knowledge can be easily implemented to assist the operator during wells’ start-up. This paper describes the integration of an expert system associated with anti-slug control for well start-up. The intelligent system has many heuristic rules to implement the automation of the start-up procedures, like the opening choke valve while simultaneously respecting many constraints. Severe slugging flow regimes are one of the major disturbances for the operation of offshore production platforms, and can cause many unplanned shutdowns. Therefore, it’s important to combine start-up intelligent system with an anti-slug advanced control module for each well. The benefits are associated to reducing possibility of unplanned shutdowns during well start-up operational procedures, decreasing operators’ stress and also helping to minimizing impacts to the environment. A prototype was implemented in one platform with good results for a safe and efficient wells start-up procedure. This paper will present the development and results of this intelligent system for wells’ start-up and anti-slug control for offshore platforms.

Downhole pressure is an important process variable in the operation of gas-lifted oil wells. The device installed in order to measure this variable is often called a Permanent Downhole Gauge (PDG). Replacing a faulty PDG is often not economically viable and to have an alternative estimate of the downhole pressure is an important goal. Using data from operating PDGs, this paper describes a number of issues dealt with in the development of soft sensors for several deepwater gas-lifted oil wells. Some of the tested models include nonlinear polynomials, neural networks, committee machines, unscented Kalman filters and filter banks. The variety of model classes used in addition to the diversity of oil wells considered brings to light some of the key-problems that have to be faced and reveal the strengths and weaknesses of each alternative solution. A major constraint throughout the work was the use of historical data, hence no specific tests were performed at any time. The aim of this work is to discuss the procedures, pros and cons of the tested solutions and to point to possible future directions of research.