This work studies the impact of Wireless Sensor Networks (WSNs) for oil spill detection and localization in Subsea Production Systems. The case study is the Goliat FPSO, with a realistic assumption about the presence of a WSN built upon the existing passive acoustic sensors installed on each subsea template to monitor the manifold. The sensors take local binary decisions regarding the presence/absence of a spill by performing an energy test. A Fusion Center (FC) collects such local decisions and provides a more reliable global binary decision. The Counting Rule (CR) and a modified Chair-Varshney Rule (MCVR) are compared. An objective function based on the Receiver Operating Characteristic (ROC) is used for threshold design. The FC, in case of a spill detection, provides an estimated position of the leak source. Four localization algorithms are explored: Maximum A-Posteriori (MAP) estimation, Minimum Mean Square Error (MMSE) estimation, and two heuristic centroid-based algorithms. Detection and localization performances are assessed in comparison to the (position) Clairvoyant Chair-Varshney Rule (CVR) and to the Cramér-Rao Lower Bound (CRLB), respectively. The considered framework requires the prior knowledge of the involved subsea production system in terms of components that in case of failure would cause a leakage and their corresponding failure rates.
29 January 2024 in IEEE Internet of Things Journal
This work proposes a data fusion approach for quickest fault detection and localization within industrial plants via wireless sensor networks. Two approaches are proposed, each exploiting different network architectures. In the first approach, multiple sensors monitor a plant section and individually report their local decisions to a fusion center (FC). The FC provides a global decision after spatial aggregation of the local decisions. A post-processing center subsequently processes these global decisions in time, which performs quick detection and localization. Alternatively, the FC directly performs a spatiotemporal aggregation directed at quickest detection, together with a possible estimation of the faulty item. Both architectures are provided with a feedback system where the network’s highest hierarchical level transmits parameters to the lower levels. The two proposed approaches model the faults according to a Bayesian criterion and exploit the knowledge of the reliability model of the plant under monitoring. Moreover, adaptations of the well-known Shewhart and CUSUM charts are provided to fit the different architectures and are used for comparison purposes. Finally, the algorithms are tested via simulation on an active Oil and Gas subsea production system, and performances are provided.
13 October 2023 in IEEE Transactions on Signal and Information Processing over Networks
This work addresses the problem of detecting gas dispersions through concentration sensors with wireless transmission capabilities organized as a distributed Wireless Sensor Network (WSN). The concentration sensors in the WSN perform local sequential detection (SD) and transmit their individual decisions to the Fusion Center (FC) according to a transmission rule designed to meet the low-energy requirements of a wireless setup. The FC receives the transmissions sent by the sensors and makes a more reliable global decision by employing a SD algorithm. Two variants of the SD algorithm named Continuous Sampling Algorithm (CSA) and Decision-Triggered Sampling Algorithm (DTSA), each with its own transmission rule, are presented and compared against a fully-batch algorithm named Batch Sampling Algorithm (BSA). The CSA operates as a time-aware detector by incorporating the time of each transmission in the detection rule. The proposed framework encompasses the gas dispersion model into the FC's decision rule and leverages real-time weather measurements. The case study involves an accidental dispersion of carbon dioxide (CO 2 ). System performances are evaluated in terms of the receiver operating characteristic (ROC) curve as well as average decision delay and communication cost.
This article presents a target detection algorithm based on spatial gradient image features to enhance radar detection performance in low signal-to-noise ratio (SNR) environments. The proposed spatial gradient-based detector (SGBD) exploits the characteristic appearance patterns of targets in radar imagery without requiring specific target feature functions. The algorithm calculates intensity variations across the image and imposes a smoothing constraint to account for target features without knowing the specific target model. The discrete implementation employs a convolution kernel to approximate gradients and Laplacians alongside an iterative solution to the Euler-Lagrange equation. The performance is evaluated against standard constant false alarm rate (CFAR) detectors using Monte Carlo simulations, highlighting the advantages of the proposed approach. The simulations further address the impact of heavy-tailed clutter distributions and spatially correlated clutter, as commonly encountered in maritime scenarios. In addition, real-world maritime radar data results are shown to validate the SGBD’s effectiveness.
In this work, we present a spatio-temporal decision fusion approach aimed at performing quickest detection of faults within an Oil and Gas subsea production system. Specifically, a sensor network collectively monitors the state of different pieces of equipment and reports the collected decisions to a fusion center. Therein, a spatial aggregation is performed and a global decision is taken. Such decisions are then aggregated in time by a post-processing center, which performs quickest detection of system fault according to a Bayesian criterion which exploits change-time statistical distributions originated by system components’ datasheets. The performance of our approach is analyzed in terms of both detection- and reliability-focused metrics, with a focus on (fast & inspection-cost-limited) leak detection in a real-world oil platform located in the Barents Sea.
This work studies the impact of Wireless Sensor Networks (WSNs) for oil spill detection in subsea Oil&Gas applications. The case study is the Goliat FPSO where one WSN with passive acoustic sensors is assumed to be installed on each subsea template to monitor the manifold. Sensors take local binary decisions regarding the presence/absence of a spill by performing an energy test. A Fusion Center (FC) collects such local decisions and provides a more reliable global binary decision. The Counting Rule (CR) and a modified Chair-Varshney Rule (MCVR) are compared. An objective function derived from the Receiver Operating Characteristic (ROC) is used for threshold design. The considered methodology requires the knowledge of the involved subsea production system, in particular of its hotspots whose failure could cause an oil spill.