A seismo-tectonic map depicting the principal structural elements of the northern Central American region has been meticulously crafted to characterize the tectonic setting and the individual seismo-tectonic structures of this area. This region is subject to heightened seismic activity, with a large number of medium-high Magnitude occurrences transpiring annually. This map is presented alongside an informative dataset wherein fault trace locations, geometry and kinematics descriptors and other available metadata have been stored. Therefore, the map offers a detailed and up-to-date depiction of the brittle deformation across the region, serving as a valuable resource for a comprehensive assessment of the seismo-tectonic framework. Moreover, the map and its accompanying database summarize fault characteristics for seismic hazard analysts and for civil protection workers, proving to be useful instruments in pinpointing areas where urgent fault research should be conducted from a seismic risk standpoint.

Earth system predictions, from sub-seasonal to seasonal timescales, remain a challenging task, and the representation of predictability sources on seasonal timescales is a complex work. Nonetheless, advances in technology and science have been making continuous progress in seasonal forecasting. In a previous paper, a performance for temperature prediction by a modelling system named e-kmf® was carried out in comparison with observations and climatology for a year of data; a low level of predictability in the sub-seasonal range, particularly in the second month, was observed over the Italian peninsula. Therefore, in this study, we focus our investigations specifically on the performance between the fifth and the eighth week of temperature forecasts over six years of simulations (2012–2018) to investigate the capability of the weather model to better reproduce the behavior of temperatures in the second month of the forecast. Although some differences in seasons are present, results have globally shown how temperature predictions have the potential to be quite skillful, with an average skill score of about 68%, with climatology used as reference; additionally, an overall anomaly correlation coefficient equal to 0.51 was shown, providing useful information for applications in planning, sales, and supply of natural energy resources.

Forecasting applications based on hourly meteorological predictions for weather variables are nowadays used in energy market operations, planning of gas and power supply, and renewable energy, among others. Available meteorological and climatological data, as well as critical thresholds of rainfall, may also have a key role in the hazard classification, related to slope instabilities of pipelines and critical infrastructures along routes. The present study concerns the performance of a weather forecast model in the framework of an early warning system (EWS) application, which supports the integrity management of oil and gas pipelines. This EWS has been applied on to a specific area: the Val d’Agri basin in the Basilicata region of Southern Italy, which is extensively affected by several landslides and floods. The hourly precipitation forecasts are provided by a dedicated meteorological model, the KALM-HD, using two different horizontal resolutions, 1.25 and 5 km, to analyze possible influences of the mesh grid size as well. On this area, several weather stations were specifically deployed to obtain observed data in a region where hydrogeological hazards are relevant for asset management. A comparison among observations and the KALM-HD scaled forecasts on six of these weather stations is presented to assess the model performance. Besides, precipitation, temperature, and wind speed are evaluated as well. The forecasting analysis is performed considering two years of data both on an overall and seasonal basis. Results show that the KALM-HD performs well with the 1.25 km grid, particularly on temperature and wind speed variables. Since weather stations can be gathered in two main sets depending on their positions, differences arise in the forecast quality of these two groups, related to orography and thermal effects, whose detection is difficult in the typical narrow valleys characterizing the area of study. This issue prevalently influences temperatures and local winds, which, these latter, are generally underestimated, while precipitation is mainly driven by synoptic circulation and its interaction with mesoscale meteorological features.

The paper addresses the integrity monitoring of critical infrastructures as fluid filled transportation pipelines by developing a pilot full scale application of Advanced Leak Detection (ALD) monitoring system. The objective of the ALD system is to improve performances with respect to single separate LD system, in different scenarios in terms of sensitivity, localization accuracy, and robustness. Three Leak Detection methods have been evaluated which rely on different physical principles, though independent, and could interact with one another through cross-links data at core level. The Advanced Leak Detection system has been developed over an already existing technological platform based on vibroacoustic sensing, which includes flowmeters, pressure sensors, and accelerometers using acquisition units distributed along the pipeline routing, and processing computers with software suite. Leak Detection sub-systems are based on i) Advanced Negative Pressure Wave for detection and localization of fluid transients; ii) RTTM-Compensated Mass Balance for detection of imbalance in the fluid mass transportation; iii) Acoustic Noise for detection and localization of existing or slow-opening leaks. The integration of different Leak Detection sub-systems can take place at various levels: cross-check of measurements, cross-link of computed fluid-dynamic quantities, merge of alarms associated to the same physical event. The Advanced Leak Detection pilot system has been deployed and validated on liquid fuel transportation pipeline 40 km long, managed by Eni SpA in North Italy. Several controlled spill-tests (i.e., with distinct size, area, shape), including spillages simulating both short-duration and slow-opening cracks, have been performed in both pumping and shut-in operational conditions to calibrate the ALD system and to assess the performances. The amount of spilled product has been measured for each test by means of a calibrated weight scale. The ALD system proved to be able to detect and localize both quick and slow-opening leaks with a good sensitivity, specificity, and precision. In addition, Leak Detection alarms contain estimates of outflow rate and hole size: accuracy and alarm response time are assessed in the paper. In conclusion, the ALD system has been validated on fuel transportation pipeline by Eni as compliant with the technical requirements and is currently deployed in operations. The novelty of the pilot ALD system lies in the integration of Leak Detection methods at the core level, that allow to exchange information, calibration and synergically contribute to provide robust, sensitive, accurate and informative Leak Detection alarms.

Pipeline transportation of multiphase products, such as gas and oil mixtures, exhibits complex and varying flow regimes: as a result, analytical approaches or conventional methods cannot accurately describe the composition or the propagation characteristics of the fluid mix inside the transportation system itself. We address such an issue by presenting a methodology, driven by the data and applied to a real case history, where basic pressure transients are used to tag and track, along a pipeline, different batches of a multiphase medium. Several statistical indicators, computed from the pressure data and on different window lengths, are employed to train a machine learning model, which learns to distinguish the characteristic behavior of two different oil-gas slugs: in practice, each different combination of fluid phases (in terms of gas/oil ratio in a given batch of product) and each different sequence of slugs (in terms of gas/oil ratio variability between successive batches) behaves like a coded tag linked to the flowing fluid. The key innovation consists in the possibility of tracking such multiphase slugs along the flowline and at each monitoring station: this allows one to determine in real-time the fluid composition entering/exiting the line, its position, and its movement along the pipe. As such, we obtain also a virtual metering system, able to provide estimates of the flow rate and phases ratio. Moreover, by having several recording stations accurately synchronized, one can also leverage real-time transmission and multichannel processing of the data, enabling the opportunity for online monitoring applications. The results on the test cases and the accuracy scores obtained for the metrics considered validate the tagging and tracking approach.

The detection of leaks in pipeline transportation systems is a matter of serious concern for operators, who pursue the integrity of their assets, the reduction of losses and the prevention of environmental hazards. Whenever a hole occurs in a pressurized pipeline, the corresponding fluid leakage is characterized by a turbulent flow and a peculiar acoustic noise, whose characteristics depend also on the size of the hole itself. This study shows that both the presence and the size of such a leaking hole can be successfully detected, by exploiting the acoustic noise (pressure transients) generated by the fluid exiting the pipe and recorded internally by hydrophones, or by considering the corresponding vibrations (e.g., acceleration signals) propagating along the external shell of the conduit. To this purpose, several experimental campaigns of acoustic noise generation have been performed using multiple calibrated nozzles on a 16” ID connection pipeline in a fuel tanks area. Detection and classification procedures are proposed to control the presence of leakages and to estimate the size of the hole, using pressure and vibration signals.

The formation of deposits is a very common issue in oil and gas pipeline transportation systems. Such sediments, mainly wax and paraffine for crude oil, or hydrates and water for gas, progressively reduce the free cross-sectional area of the pipe, leading in some cases to the complete occlusion of the conduit. The overall result is a decrease in the transportation performance, with negative economic, environmental, and safety consequences. To prevent this issue, the amount of inner deposits must be continuously and accurately monitored, such that the corresponding cleaning procedures can be performed when necessary. Currently, the former operation is still dictated by best-practice rules pertaining to preventive or reactive approaches, yet the demand from the industry is for predictive solutions that can be deployed online for real -time monitoring applications. The paper moves toward this direction by presenting a machine learning methodology that leverages pressure measurements to perform online monitoring of the inner deposits in crude oil trunklines. The key point is that the attenuation of pressure transients within the fluid is dependent on the free cross-sectional area of the pipe. Pressure signals, collected from two or more distinct locations along a pipeline, can therefore be exploited to estimate and track in real time the presence and thickness of the deposits. Several statistical indicators, derived from the attenuation of such pressure transients between adjacent acquisition points, are fed to a data-driven regression algorithm that automatically outputs a numeric indicator representing the amount of inner pipe debris. The procedure is applied to the pressure measurements collected for one and a half years on discrete points at a relative distance of 40 and 60 km along an oil pipeline in Italy (100 km length, 16 -in. inner diameter pipes). The availability of historical data prepipe and postpipe cleaning campaigns further enriches the proposed data-driven approach. Experimental results demonstrate that the proposed predictive monitoring strategy is capable of tracking the conditions of the entire conduit and of individual pipeline sections, thus determining which portion of the line is subject to the highest occlusion levels. In addition, our methodology allows for real -time acquisition and processing of data, thus enabling the opportunity for online monitoring. Prediction accuracy is assessed by evaluating the typical metrics used in the statistical analysis of regression problems.

In oil depots and fuel storage facilities, undetected storage tanks damages can lead to the leakage of the oil stored in the soil leading to pollution and economical losses. Leaks are generally due to the perforation of the storage tank floor due to corrosion. The detection of corrosion and leaks is a complicated task, especially for operative tanks with inaccessible floor for detailed inspections and is generally attempted by mean of acoustic emission systems operating from the outer skin of the tank. In this paper, we present a compact sensor node (SN) designed for long-term and real-time acoustic emission monitoring. The SN exploits up to three inexpensive low-frequency sensors based on piezoelectric diaphragms, and it is capable by means of built-in Digital Signal Processing functionalities to process the acquired time waveforms extracting the AE features usually required by testing protocols. An experimental validation on a floating-roof aboveground storage tank 17 m high and 18 m in diameter, filled with water to a level of about 6.2 m, is proposed. Leaks were induced by opening and closing a drainage valve existing at the bottom skirt of the storage tank while acoustic emission signals were recorded at three sensors and processed in real time. Designed nozzles of different diameter, from 1 mm to 9 mm, where used to simulate leakages of different entities. The results confirm the possibility of detecting and monitoring leaks of various diameters in the low-frequency region 1–2 kHz not traditionally considered by state-of-art acoustic-emission monitoring systems.

Pumping systems are a key component of oil and gas pipeline transportation assets: monitoring their integrity is a crucial operation from a safety and revenue point of view. The solutions currently employed in the industry apply supervised machine learning techniques to data collected by multi-domain sensors directly installed on several positions of the pump itself; however, such approaches are not applicable on older machines, in contexts where a direct access to the pump is not possible, or whenever labelled data are not at disposal. This paper, instead, presents a predictive maintenance strategy where the condition of a centrifugal pump is tracked by solely exploiting standard pressure measurements, recorded also on remote points along the pipeline, and using an unsupervised learning approach. The smart monitoring strategy is presented and validated on historical pressure signals collected by Eni for several years on a crude oil transportation pipeline, located in Italy. Pressure data, recorded along the fluid line, are used to compute several statistical indicators on appropriate window lengths. These indicators are then fed to an unsupervised clustering procedure, based on a Gaussian mixture model. The output is an index within four different pump operational regimes, and a clustering visualization that permits the interpretation of the automatic regime classification. In fact, the manual inspection of the clusters shows that three of them describe standard modes (regular pumping operation, pumps off, flow regulations). The fourth one corresponds to high amplitude peaks in the signals and indicators, and so it is tagged as "anomalous" mode: pump maintenance logs reveal that the peaks are associated to damaged roller bearing movements, which disappear after the activation of the pump backup system. Anomalies are reported several days before the pump switch, so that a preventive maintenance could have been triggered. The robustness of the clustering algorithm is assessed on a statistical basis, whereas the overall validity of the monitoring system is tested on an instantaneous basis by applying the proposed model on two independent datasets, collected on real transportation pipelines: the results demonstrate the reliability of the proposed monitoring strategy in predicting and detecting all the pump failure events reported by the available maintenance logs. With respect to the mostly employed approaches, our machine learning procedure does not require any previous supervision of the data. Moreover, input data are the pressure transients produced by the pumps and guided within the fluid in the pipeline for long distances: pump failure analysis can be run using sensors located at several kilometers of distance from the pump itself, making a remote control strategy feasible.

This paper presents an innovative machine learning methodology that leverages on long-term vibroacoustic measurements to perform automated predictions of the needed pigging operations in crude oil trunklines. Historical pressure signals have been collected by Eni (e-vpms monitoring system) for two years on discrete points at a relative distance of 30-35 km along an oil pipeline (100 km length, 16 inch diameter pipes) located in Northern Italy. In order to speed up the activity and to check the operation logs, a tool has been implemented to automatically highlight the historical pig operations performed on the line. Such a tool is capable of detecting, in the observed pressure measurements, the acoustic noise generated by the travelling pig. All the data sets have been reanalyzed and exploited by using field data validations to guide a decision tree regressor (DTR). Several statistical indicators, computed from pressure head loss between line segments, are fed to the DTR, which automatically outputs probability values indicating the possible need for pigging the pipeline. The procedure is applied to the vibroacoustic signals of each pair of consecutive monitoring stations, such that the proposed predictive maintenance strategy is capable of tracking the conditions of individual pipeline sections, thus determining which portion of the conduit is subject to the highest occlusion levels in order to optimize the clean-up operations. Prediction accuracy is assessed by evaluating the typical metrics used in statistical analysis of regression problems, such as the Root Mean Squared Error (RMSE).

Abstract In recent years, big data technologies have paved the way for digital transformation in oil and gas industry. Multi-domain measurements are collected by advanced sensor systems and processed using data-driven approaches, allowing to derive constitutive relations between the operational status of the asset and the measured variables. In addition, historical pressure measurements can be exploited for advanced pipeline monitoring. This paper presents a methodology, applied to a case history, where legacy data are repurposed and employed both to track pump health and to enhance the digital conversion. The dataset consists of past pressure signals collected by Eni for several years at the pumping terminal of a crude oil transportation pipeline, which has a length of 100 km and 16" diameter pipes, located in Italy. Pressure transients' variance, kurtosis and variation range, computed on appropriate window lengths, are fed to an unsupervised clustering procedure based on a Gaussian Mixture Model (GMM), which automatically identifies four clusters. An expert analysis of the labeled data reveals that each cluster corresponds to a well-defined and different pump operational mode, namely: standby (pumps off), transition (pumps switching on/off), normal (line flowing) and anomalous. The latter mode is connected to a high value in the pressure transients' variance and kurtosis: during such regime, pump maintenance logs report a failure and replacement of a system part. Interestingly, the anomalous condition starts to show up several days before the actual part replacement. The proposed case history reveals the potentiality of: adding value to legacy data, as they can be reprocessed, tagged and used as supervised examples in the training phase of new data-driven procedures; comparing, merging and complementing monitoring strategies of assets at different digitalization stages; aiding the development of predictive maintenance strategies.

Pipeline Inspection Gauges (PIGs) are widely used for monitoring and managing pipeline integrity. During a pigging operation it is fundamental to have a continuous measurement of the PIG position and movement, in order to achieve the best inspection and to have an early warning in the case the device is stuck. Currently, the tracking is performed by installing on the PIG "active" systems (e.g. acoustic pingers or electromagnetic emitters) that communicate with a set of receivers, making it possible the localization of the travelling gauge. Another solution is the deployment of an appropriately dense network of sensors along the pipe track, or the utilization of a dedicated system/crew that moves close to the pipe, so to physically perceive the vibrations generated by the nearby passage of the device. In fact, the moving PIG produces pressure transients and vibro-acoustic noise due to the velocity fluctuations, to the friction against the internal walls and to the crossing of the welding dents. It is important to mention that the conduit acts like an acoustic waveguide and the "sound" generated by the PIG, in many practical situations, can be sensed within the fluid at several kilometres from the originating point. This paper presents three different tracking procedures that exploit the noise generated by the PIG to locate it, remotely and passively, without requiring any additional equipment to be mounted on the gauge. The key points of the procedures are the availability of pressure measurements at a small number of positions along the pipeline, at relative distances of tens of kilometres, an accurate synchronization of the measurements, the real time transmission and multichannel processing of the data. The first method locates the PIG by performing a crosscorrelation analysis between the acoustic signal recorded on opposite sides of the moving gauge, the second method is based on the counting of the transients generated at known positions, the third one describes the pipe section between the PIG and the arrival terminal like a resonant structure, and obtains the length of this section (the distance of the PIG to the arrival) from the resonance frequency. All the methods are presented starting from real examples, in order to highlight their effective applicability. Moreover, the localization results are in agreement with the output of more sophisticated technological solutions.

A short-term forecast of energy consumption is affected by different factors related to the demand in residential, commercial, thermoelectric, and industrial sectors. This demand can be strongly constrained by weather variables, especially temperatures, whose forecast may be very useful to predict the balances between supply and demand, minimizing the risk of price volatility. Energy companies use the relationship between meteorological forecast output and energy request to provide an effective scheduling of national gas and power grids and reduce operational costs in critical periods. This work reports a comparison analysis for short- and medium-term daily temperature forecasts during the period 2013-2014 by using the weather model e-kmf™ (eni-kassandra meteo forecast), currently adopted in gas and power applications where meteorological output has a key role. This weather forecast system uses different models and initial data to develop probabilistic predictions from a perspective of eleven days ahead. In particular, a set of model runs with horizontal grid spacing of 5.5, 8, 13, and 18 km with the same domain size are undertaken to assess the sensitivity of temperature to horizontal resolutions. A nonlinear Kalman filter has been also applied to postprocess forecasted data in eight European cities (Milano, Roma, Torino, Napoli, Munich, Paris, Brussels, and London). Filtered forecasts over these cities have been compared to local observations taken from SYNOP (surface synoptic observations) and METAR (meteorological Aerodrome Report) stations. Skill scores of performance have been used to generally assess the forecast reliability up to day +11. In order to understand the sensitivity to the horizontal resolution, investigations have been carried out even during four specific periods of two weeks with stable and unstable weather conditions.

Efficiency and safety are primary requirements for oil & gas fluid filled transportation system. However, the complexity of the asset makes it challenging to derive a theoretical framework for managing the control parameters. The current frontier for a real time monitoring exploits the "digital tansformation", i.e. the acquisition and the analysis of large datasets recorded along the whole asset lifecycle, which are used to infer "data driven" relations and to predict the evolution of the asset integrity. This paper presents some results of a research project for the design, implementation and testing of a "machine learning" approach to vibroacoustic data recorded continuously by acquisition units installed every 10-20 km along a pipeline. In a fluid transportation system, vibroacoustic signals are generated by the flow regulation equipment (i.e. pumping, valves, metering), by the fluid flowing (i.e. turbulence, cavitation, bubbles), by third party interference (i.e. spillage, sabotage, illegal tapping), by internal inspection using PIGs operations), and by natural hazards (i.e. microseismic, subsidence, landslides). The basic principle of machine learning is to "observe", for an appropriate time interval, a series of descriptors, in this stage related to vibroacoustic signals but that can be integrated with other physical data (i.e. temperature, density, viscosity), in order to "learn" their safe range of variation or, when properly fed to a classification procedure, to obtain automatically a discrete set of operational status. The classification criteria are then applied to new data, highlighting the presence of system anomalies. The paper considers vibroacoustic signals collected at the flow stations of an oil trunkline in Nigeria. The vibroacoustic signals are the static pressure, the acceleration and the pressure transients recorded at the departure and at the arrival terminals. More than one year of data is available. Derived smart indicators are defined, which are directly linked to the asset parameters: for instance, the cross-correlation of the pressure transients at adjacent measuring locations permits to estimate the fluid channel continuity (correlation value), the sound velocity (time of correlation peak), and the sound attenuation (amplitude versus frequency amplitude decay). A portion of the data during normal operation is used for training and tuning a reference model. After that, new data are compared with the model, and anomalies are automatically detected. Two kind of errors are raised: i) sensors; ii) alerts. Sensor errors are referred to missing or corrupted sensors data. Alerts are raised when the measured physical quantities are not coherent with the functional and known service behaviors of the transport system. The system model is not static over time, and in fact it can be updated by the operators’ feedback, that can tag false alarms and thus, automatically, re-define the set of operational scenarios of the upstream system. The medium-long term construction and update of data driven models is effective for predictive maintenance, automatic anomalies detection, optimization of operational procedures. Moreover, the new policy of data management and the opportunity of gaining awareness by interconnecting the monitoring experience of different assets leverages the introduction of new technologies (cloud, big data), new professional figures (smart data scientist), new operational and business models.

The paper proposes a method to estimate 2D/3D vibrations and displacements of mostly linear structures, like pipes, chimneys, towers, bridges from afar, based on synchronized Radars. The method takes advantage of Radar sensitivity to displacements to sense tiny deformations (up to tens of micron) with a time scale from milliseconds to hours. The key elements are: (a) The use of calibrators to remove at once both the tropospheric turbulence and the effect of radial motion, and (b) the compensation of interferences from fixed targets. The latter is performed by estimating and removing the contribution of interfering targets, based either on a proper data processing or by exploiting an ad-hoc motorized calibrator. Performance in terms of accuracy of the deformation field is evaluated theoretically and checked by tests carried out in laboratories and by full-scale acquisition campaigns.

An hourly short-term weather forecast can optimize processes in Combined Cycle Gas Turbine (CCGT) plants by helping to reduce imbalance charges on the national power grid. Consequently, a reliable meteorological prediction for a given power plant is crucial for obtaining competitive prices for the electric market, better planning and stock management, sales and supplies of energy sources. The paper discusses the short-term hourly temperature forecasts, at lead time day+1 and day+2, over a period of thirteen months in 2012 and 2013 for six Italian CCGT power plants of 390 MW each (260 MW from the gas turbine and 130 MW from the steam turbine). These CCGT plants are placed in three different Italian climate areas: the Po Valley, the Adriatic coast, and the North Tyrrhenian coast. The meteorological model applied in this study is the eni-Kassandra Meteo Forecast (e-kmfTM), a multi-model approach system to provide probabilistic forecasts with a Kalman filter used to improve accuracy of local temperature predictions. Performance skill scores, computed by the output data of the meteorological model, are compared with local observations, and used to evaluate forecast reliability. In the study, the approach has shown good overall scores encompassing more than 50,000 hourly temperature values. Some differences from one site to another, due to local meteorological phenomena, can affect the short-term forecast performance, with consequent impacts on gas-to-power production and related negative imbalances. For operational application of the methodology in CCGT power plant, the benefits and limits have been successfully identified.

Integrity and operational reliability of pipelines requires a continuous monitoring, able to detect, in real time, failure, sabotage and leak of fluids (e.g. natural gas, crude oil, water, products), and to protect the safety of personnel and local communities. In the framework of a research project, eni has developed a proprietary pipeline monitoring technology based on vibroacoustic (negative pressure wave) sensing. This paper presents some experimental and theoretical activities for the evaluation of the system performance versus fluid leaks, in terms of maximum detection distance and sensitivity. The field scenario is a 12 km, 3 inch inner diameter sealine in south of Italy, conveying a fluxing agent from an onshore terminal to an offshore platform. Leak-like events have been produced by opening discharge valves at the terminals. The associated travelling pressure transient, recorded by several monitoring stations, has been compared with the environmental noise (generated by the flow regulation equipment) and with the theoretical models. The predicted propagation parameters and maximum detection distance are in very good agreement with the measured ones: this allows us to use the theoretical simulator for design and performance evaluation of new installations.

In this paper we present a novel method for daily short-term load forecasting, belonging to the class of "similar shape" algorithms. In the proposed method, a number of parameters are optimally tuned via a multi-objective strategy that minimizes the error and the variance of the error, with the objective of providing a final forecast that is at the same time accurate and reliable. We extensively compare our algorithm with other state-of-the-art methods. In particular, we apply our approach upon publicly available data and show that the same algorithm accurately forecasts the load of countries characterized by different size, different weather conditions, and generally different electrical load profiles, in an unsupervised manner.

This paper illustrates and compares the ability of several clustering algorithms to correctly associate a given aggregate daily electrical load curve with its corresponding day of the week. In particular, popular clustering algorithms like the Fuzzy c-Means, Spectral Clustering and Expectation Maximization are compared, and it is shown that the best results are obtained if the daily data are compressed with respect to a single feature, namely the so-called “Morning Slope”. Such a feature-based clustering appears to outperform the clustering results obtained upon using other classic features, and also with respect to using other conventional compression methods, such as the Principal Component Analysis, in all the examined European countries. This result is particularly interesting, as this feature provides a direct physical interpretation that can be used to obtain insights on the structure of the daily load profiles.