Giuliani, Marco (Eni S.p.a) | Cadei, Luca (Eni S.p.a) | Montini, Marco (Eni S.p.a) | Bianco, Amalia (Eni S.p.a) | Niccolai, Alessandro (Politecnico di Milano) | Mussetta, Marco (Politecnico di Milano) | Grimaccia, Francesco (Politecnico di Milano)
Asset optimization has recently become a crucial issue in Oil&Gas industry, considering oil price conjuncture and an increased awareness on environmental aspects. In this paper, an Artificial Intelligence (AI) technique is presented, which is able to manage big dataset to automatically match the entire production model against measured field data. The tool is based on a hybrid in-house developed AI technique, integrating deep neural networks, biogenetical algorithms, commercial simulators and real-time data.
The workflow starts with the modeling of the production system through physics-based commercial simulators. A sensitivity analysis identifies the critical variables, which are then randomly varied with a Sobol distribution, exploring the entire solution domain. With these data, a proxy model to the commercial software is generated using an artificial neural network. Finally, the AI tool fed by real-time data is used to match the field behavior: uncertain parameters are modified through a differential evolution algorithm that minimizes the error between calculated and measured variables. The matching parameters are, then, passed to the simulators achieving a field representative model.
The tool has been developed considering an operating field in offshore western Africa. The typical uncertain parameters in this kind of field are related to the fluid characteristics, in particular densities and compositions, but also to the physical characterization of the pipelines such as roughness and heat transfer characteristics. The matching process has been performed coupling the proxy model, which is a neural network able to replicate the field behavior, and a differential evolution algorithm as the optimization algorithm. The fitness function to be minimized is a Mean Absolute Percentage Error (MAPE) that represents the distance between the actual field production parameters and the modelled ones. The best configuration of both the neural network and the differential evolution algorithm required a computational time of 6 seconds with a MAPE equal to 2.6%. These results are compared to the one obtained coupling the same differential evolution algorithm with the commercial simulator to perform the matching. The required computational time is equal to about 20 hours (70400s) and a MAPE equal to 2.2%. The big gain with the novel approach is clearly the knocking down of computational time with a comparable error.
In this paper, it has been shown how substituting the physical model with a proxy one can give substantial advantages in terms of computational time. In principle, with the velocity of the tool implemented, the matching procedure could be done on a daily basis. This is a breakthrough because it allows having the simulator model always tuned and ready to be utilized.
Although, with the usual seismic acquisition, much of the information about subsurface parameters is contained in the reflections data, many of the current Full Waveform Inversion applications still focus on the use of the transmitted energy (i.e. refractions). In order to reduce the dependency on refractions, in this work we propose to pre-process the recorded and the simulated signals with an exponential function in order to be able to exploit the Normalized Integration Method also for reflection data.
Tests on synthetic data shows that the resulting new cost function is more robust with respect to the cycle skip compared to the standard least squares cost function. We show the results of the proposed method on two simple synthetic datasets and on a more realistic one characterized by complex geology and strong velocity contrasts
Presentation Date: Thursday, October 18, 2018
Start Time: 8:30:00 AM
Location: 207C (Anaheim Convention Center)
Presentation Type: Oral
Prando, Davide (Politecnico di Milano) | Bolzoni, Fabio (Politecnico di Milano) | Nicolis, Davide (Politecnico di Milano) | Pedeferri, MariaPia (Politecnico di Milano) | Ormellese, Marco (Politecnico di Milano)
The corrosion behavior of commercially pure titanium (UNS R50400, ASTM GRADE 2) was investigated in presence of aggressive, bromides containing solution reported to cause more severe localized corrosion compared to chlorides. To enhance localized corrosion resistance of the metal, chemical oxidation treatments were performed using NaOH 10 M solution at room temperature and at 60°C. Treatment duration effect on final corrosion resistance of samples was investigated spanning from 1 h to 72 h. After treatment optimization, the best one was compared to anodic oxidation at low potential.
To further increase corrosion resistance, annealing at 400°C and 600°C was performed after chemical oxidation and the resulting samples were tested in bromides containing solution.
Titanium has outstanding corrosion resistance due to a thin, amorphous, non-stoichiometric TiO2 protective layer (max 10 nm thick1) that is formed spontaneously on the surface when this is exposed to an aerated environment. This protective layer is very stable and allows the use of titanium in severe working conditions, such as offshore (up to 260°C), acid environments, aerospace2,3, automotive, high temperature, chemical & food industry, 4-6 marine hydrometallurgical applications and nuclear fuel wastes containment, 7,8 where no other metal can be used.
Nevertheless, commercially pure titanium may suffer different forms of corrosion in very severe environments.9 Generalized corrosion is caused by small quantity of fluorides ions (more than 0.002 M) that combining with titanium forming TiF4, destroying passivity film. Hydrogen embrittlement happens prevalently on α- and α-β titanium due to their low hydrogen solubility in α-Ti, stress corrosion cracking can also happen in very specific environments, such as anhydrous methanol, nitrogen tetroxide, red- fuming nitric acid or solid cadmium.
However, the most critical forms of corrosion of titanium are due to localized breaking of passive layer and this is favored by the presence of concentrated halides, such hot salty water (above 200°C) or bromide containing species.
Brenna, Andrea (Politecnico di Milano) | Brugnetti, Fabio (Snam Rete Gas S. p. A.) | Refraschini, Beatrice (Politecnico di Milano) | Beretta, Silvia (Politecnico di Milano) | Ormellese, Marco (Politecnico di Milano)
The Italian ministerial decree (D.M. 4 April 2014) foresees that at rail crossings and parallelisms, pipes for liquid and gas transportation shall be encased in a well-sealed, coated steel pipe called casing pipe. Laboratory tests have been carried out to investigate the protection and interference condition of both the carrier pipe and the casing in the case where water is present in the annulus. Cathodic protection was applied to the gas pipe by an impressed current system, cell voltage was increased and the IR-free potential of both pipes was measured. Different configurations were considered for the casing: bare tube with and without a through hole; coated tube with a defect on the internal and external surface; bare tube electrically connected to the internal pipe by a shunt. In the presence of a perfect watertight seal, no interference takes place on the casing pipe, as expected. If an electrolyte is present in the annulus, overprotection conditions on the internal pipe are reached in all tested conditions, except for test performed on the connected pipes. In the presence of a bare casing, interference effects are negligible but overprotection condition on the internal tube is reached at the lowest cell voltage.
Stray currents originating from direct current (d.c.) systems may cause severe material damage by corrosion on buried or immersed metal structures1. To prevent the effects of stray current corrosion caused by direct current, different methods can be used, such as cathodic protection (c.p.), forced drainage bond or insulating joints1,2. The Italian ministerial decree D.M. 4 Aprile 20143 demands that ducts buried in correspondence of rail crossings or parallelisms shall be encased in a coated steel pipe, called casing pipe. To ensure the protection of the inner tubing (carrier pipe), the ends of the casing pipe must guarantee the perfect watertight seal of the annulus while spacers of insulating material keep the two tubes electrically separated (Figure 1). In practice, it is almost impossible to achieve a perfect watertight seal. To avoid corrosion of the carrier pipe, a first possibility is the injection of a suitable filler material into the annular space. The filler material should either inhibit corrosion (e.g. visco-elastic compounds, inhibited wax) or be designed to allow c.p. current to reach the carrier pipe4. For casings that pass c.p. current (i.e. bare or poorly coated steel pipes or uncoated concrete pipes), the external cathodic protection of the carrier pipe can be effective in protecting the carrier pipe provided there is no contact between the carrier pipe and the casing, and that there is enough electrolyte in the annular space. Without any electrolyte in the annular space, atmospheric corrosion can occur at coating defects. Moreover, if c.p. of the casing is required, the casing should be resistively bonded to the carrier pipeline. The aim of this work is to investigate the behavior of the pipes in cases when the insulation of the carrier tube is not granted: this can occur if an electrolyte is present between the two tubes because of a leakage in the watertight system or in the casing. Laboratory tests have been carried out to investigate the protection (and over-protection) of the carrier pipe (from here on called ”gas pipe”) and the interference condition of the casing pipe, in case of water in the annulus. Different conditions were taken into consideration in order to verify the effects of coatings on the casing pipe and eventually the effect of an electrical connection between the two tubes, which aims at avoiding anodic interference of the internal surface of the casing tube. In all conditions, c.p. was applied to the gas pipe and both tubes were submerged in a conductive solution. This way it was possible to analyze the level of protection (or overprotection) on the gas pipe and the conditions of anodic/cathodic interference on the casing pipe.
Corrosion inhibitors are used to prevent or delay corrosion in reinforced concrete structures. Available commercial products are based on nitrite or organic mixtures based on amines, alkanolamines, fatty acids and carboxylic substances. During the last 15 years, an intense experimental research has been carried out focused at identifying new organic substances or binary mixtures that might have inhibiting effectiveness on chloride-induced corrosion. The paper describes results of potentiodynamic and potentiostatic electrochemical tests carried out in alkaline solution, in the presence of chlorides, on binary mixtures with nitrite, dimethylethanolamine (DMEA), glutamine and benzoate. Some of the mixtures exhibited a marked synergistic effect in the potentiodynamic polarisation tests, with a clear increase in pitting potential. A certain effect on chloride threshold was also observed in potentiostatic polarisation. The best mixtures are able to increase the critical chloride content, then accordingly they can delay corrosion initiation on real reinforced concrete structures.
Durability of reinforced concrete structures depends on carbon steel rebars corrosion. The two main causes of corrosion are concrete carbonation, due to the ingress of CO2, which reduces concrete pore solution pH to 8, and ingress of chlorides.1 In the latter case, corrosion occurs if at the rebar level, chloride concentration exceeds a threshold value, which is influenced by the chemical composition of the rebar, concrete pH and rebar electrochemical potential, as clearly described by the ”Pedeferri Diagram” for cathodic protection and prevention.2-3 Corrosion prevention is achieved during the design phase by making a high quality concrete mixture proportion, with a low water/cement ratio, by performing a correct curing and casting, and by using an appropriate cover. The European standards set out the threshold values of such parameters in relation to environmental aggressiveness.4-5 With regard to structures exposed to very corrosive environments, or for structures with a design life over 50 years, it would be appropriate to use additional protective methods: blended cements, corrosion- resistant reinforcements, inhibitors, concrete coatings and cathodic protection.
SUMMARY In this work we describe a machine learning pipeline for facies classification based on wireline logging measurements. The algorithm has been designed to work even with a relatively small training set and amount of features. The method is based on a gradient boosting classifier which demonstrated to be effective in such a circumstance. A key aspect of the algorithm is feature augmentation, which resulted in a significant boost in accuracy. The algorithm has been tested also through participation to the SEG machine learning contest.
Travel time tomography deals with the estimation of sub-surface parameters affecting the wave velocity through the medium. It is well-known this is a strongly ill-posed and ill-conditioned inverse problem, which exhibits great structural uncertainties and ambiguities about the true velocity model. Since the estimation of velocities significantly affects depth image analysis and drilling decisions, performing an investigation of model uncertainty becomes paramount in seismic processing workflow. The uncertainty issue is typically treated by relating the algebraic features of the forward operator, which links the acquired measurements to subsurface parameters, with its singular value decomposition. However, when facing geophysical inversions, this strategy becomes infeasible because of the large dimensioning of data. We propose an alternative approach for uncertainty analysis based on Monte Carlo methods. Specifically, we will show how it is possible to evaluate ambiguities related to tomographic inversion by randomly sampling the space of the forward operator. Our method offers a viable alternative solution to singular value decomposition for small-scale problems, at the same time allowing a complete problem characterization for large-scale inversions. In order to show the strength and effectiveness of this proposal, we report a tutorial Velocity Model Building problem, exploring the role of Monte Carlo methods and regularization for uncertainty analysis.
Presentation Date: Wednesday, September 27, 2017
Start Time: 11:00 AM
Presentation Type: ORAL
AbstractThis work is focused on exploring the applicability of intelligent methods in assessing porosity and permeability in the context of reservoir characterization. The main motivation underlying our study is that appropriate estimation of reservoir petrophysical parameters such as porosity and/or permeability is a key step for in-situ hydrocarbon reservoir evaluation. We ground our analysis on information on log-depth, caliper, conductivity, sonic logging, natural gamma, density and neutron porosity, water saturation, percentage of shale volume, and type of lithology collected from well loggings in an oil field in the middle-east (a total number of 11 exploratory wells are considered). Data also include porosities and permeabilities evaluated on core samples from the same wells. All these data are embedded in a neural network-based approach which enables us to establish input-output relationships in terms of an optimized number of input variables. Three diverse intelligent techniques are tested. These include: (i) classical artificial neural networks; (ii) artificial neural networks based on principal component analysis (PCA) transformation; and (iii) statistical neural networks based on a bagging approach. Our results suggest that the statistical neural network is most effective for the field setting considered. The application of this neural network with 9 input parameters provides reliable performances in 94% and 81% of the cases, respectively in the training and validation phases, for the estimation of porosity. A trained network with 10 input parameters leads to successfull reproduction of permeability values in 85% and 79.5% of the cases, respectively during training and validation of the network. Results from this study are expected to be transferable to applications involving evaluation of petrophysical properties of a target reservoir in the presence of incomplete well log datasets.
Fossati, Fabio (Politecnico di Milano) | Bayati, Ilmas (Politecnico di Milano) | Muggiasca, Sara (Politecnico di Milano) | Vandone, Ambra (Politecnico di Milano) | Campanardi, Gabriele (Politecnico di Milano) | Burch, Thomas (Centre Suisse d'Electronique et de Microtechnique) | Malandra, Michele (North Sails Group)
Also the project management and commissioning are described, as well as the measurement capabilities and data acquisition procedure. Furthermore, an important feature of this project is the availability of measurement systems for pressure distribution acting on the sails at full scale. In the following, the pressure measurement system is described in detail, as well as the data acquisition process and system metrological validation is provided. The pressure measurement system has also been tested in the wind tunnel using a scale model of a sailing yacht and compared with a different pressure measurement system already available at Politecnico di Milano Wind Tunnel. For wind tunnel tests strips and pads adequate for the model sails were used.
This paper presents a novel CFD analysis of an Oil& Gas separator, based on a multi-fluid Eulerian-Eulerian model of the Navier-Stokes equations, implemented in OpenFOAM®. The simulation of a three-phase separator poses a particular challenge to the numerical modeling of transport phenomena since the three-phase flow can span across multiple flow regimes from disperse to separate. To handle such complex behavior, a new three-phase Eulerian-Eulerian solver has been implemented in OpenFOAM with a fully implicit treatment of drag terms and with the capability to describe both disperse and separate flow at high, fully coupled phase fractions. Furthermore, the mixture turbulence model implemented in OpenFOAM for bubble flows has been improved. Firstly, the source term of the turbulent kinetic energy has been modified with a more regime-independent formulation derived from the literature. Then, the derivation of the same model has been extended in order to manage the three phases.
The work represents an improvement both from an academic and industrial perspective: it provides a consistent numerical framework for a multiphase flow involving a number of phases higher than two; it replaces the traditional Eulerian-Lagrangian approach with the more appropriate Eulerian-Eulerian one for the analysis of industrial production facilities. These two aspects allow to describe more accurately the flow pattern transitions and to numerically capture the separation and phase inversion phenomena inherent to the system.
Nowadays, multiphase flows can be considered as the standard condition for the majority of the Oil&Gas production fields. The reasons for this peculiarity are numerous and can be related to different parts of the hydrocarbons production chain. As an example, injection of fluids to sustain production as well as the presence of an active gas cap or an aquifer determine the existence of at least two different phases in the upcoming flow streams. Moreover, the difference in pressure between the wellbore and the production tubing determines the production of a certain (and in many cases non negligible) amount of gas (if the bubble point condition is reached) and the possible arising of non-ideal volumetric behaviors (e.g., retrograde condensation).