Expert Systems
The prevailing methodology in data-driven fault detection leverages synthetic data for training neural networks. However, it grapples with challenges when it comes to generalization in surveys exhibiting complex structures. To enhance the generalization of models trained on limited synthetic datasets to a broader range of real-world data, we introduce FaultSSL, a semi-supervised fault detection framework. This method is based on the classical mean teacher structure, in which its supervised part employs synthetic data and a few 2D labels. The unsupervised component relyies on two meticulously devised proxy tasks, allowing it to incorporate vast unlabeled field data into the training process. The two proxy tasks are PaNning Consistency (PNC) and PaTching Consistency (PTC). PNC emphasizes the feature consistency in overlapping regions between two adjacent views in predicting the model. This allows for the extension of 2D slice labels to the global seismic volume.#xD;PTC emphasizes the spatially consistent nature of faults. It ensures that the predictions for the seismic, whether made on the entire volume or on individual patches, exhibit coherence without any noticeable artifacts at the patch boundaries. While the two proxy tasks serve different objectives, they uniformly contribute to the enhancement of performance. Experiments showcase the exceptional performance of FaultSSL. In surveys where other mainstream methods fail to deliver, we present reliable, continuous, and clear detection results. FaultSSL reveals a promising approach for incorporating large volumes of field data into training and promoting model generalization across broader surveys.
- Reservoir Description and Dynamics > Reservoir Characterization > Seismic processing and interpretation (1.00)
- Data Science & Engineering Analytics > Information Management and Systems > Neural networks (1.00)
- Data Science & Engineering Analytics > Information Management and Systems > Artificial intelligence (1.00)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Diagnosis (0.93)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Expert Systems (0.83)
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks > Deep Learning (0.68)
Utilizing Artificial Intelligence and Knowledge-Based Engineering Techniques in Shipbuilding: Practical Insights and Viability
Shahzad, Tufail (School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an, China / Department of Innovation, MasterShip Software BV, Eindhoven, The Netherlands) | Wang, Peng (School of Marine Science and Technology, Northwestern Polytechnical University, Xi’an, China) | van Lith, Peter (Eindhoven University of Technology (TU/e), Eindhoven, The Netherlands) | Hoffmans, Jacques (Department of Research & Development, MasterShip Software BV, Eindhoven, The Netherlands)
_ This paper delves into the technical aspects and viability of integrating artificial intelligence (AI) and knowledge-based engineering (KBE) tools in practical design. The goal is to digitally embed the hands-on expertise and technical boundaries set by seasoned professionals during intricate engineering and preparatory phases. We showcase how AI/KBE tools might emulate human cognitive processes to make well-informed choices. The article also probes the prospective economic and modernization repercussions of this innovation. Our findings suggest that such an integration is feasible and can amplify the decision-making efficacy and advance the sophistication of CAD/CAM systems in the shipbuilding realm. Furthermore, this investigation underscores the promising future of AI/KBE tools in ship design and advocates for continued exploration and innovation in this sector to fully harness its advantages. Introduction Shipbuilding has long been intertwined with CAD/CAM technologies. As technology evolves, so does the landscape of ship design and manufacturing (Ross, 1950). Traditionally, ship design leaned heavily on seasoned engineers and designers, whose insights were cultivated over years of experience. However, with the rising demand for ships and an aging workforce, there’s a pressing need for enhanced design methodologies. Enter the era of artificial intelligence (AI) and knowledge-based engineering (KBE), which promise to revolutionize ship design by integrating practical knowledge and technical constraints. In today’s shipbuilding scenario, younger engineers often handle detailed engineering stages, a shift from when experienced professionals dominated the shop floor (Moyst and Das, 2005). Our research aims to assess the feasibility of AI KBE systems in enhancing the ship design process during these stages, by virtualizing the knowledge of experienced workers.
- Research Report > New Finding (0.86)
- Overview (0.66)
- Transportation > Marine (1.00)
- Shipbuilding (1.00)
- Information Technology > Knowledge Management > Knowledge Engineering (1.00)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Expert Systems (1.00)
The Monitoring of Abnormal Fluid Properties Based on PCA Technique as an Alternative Strategy to Support Autonomous Drilling Operations
Borges Filho, Moacyr N. (Programa de Engenharia de Processos Químicos e Bioquímicos/EQ, Universidade Federal do Rio de Janeiro (Corresponding author)) | Mello, Thalles (Programa de Engenharia de Processos Químicos e Bioquímicos/EQ, Universidade Federal do Rio de Janeiro) | Scheid, Claudia M. (Departamento de Engenharia Química, Universidade Federal Rural do Rio de Janeiro) | Calçada, Luis A. (Departamento de Engenharia Química, Universidade Federal Rural do Rio de Janeiro) | Waldmann, A. T. (Centro de Pesquisas, Desenvolvimento e Inovação Leopoldo Américo Miguez de Mello–CENPES/PETROBRAS) | Martins, André Leibsohn (Centro de Pesquisas, Desenvolvimento e Inovação Leopoldo Américo Miguez de Mello–CENPES/PETROBRAS) | Pinto, José C. (Programa de Engenharia de Processos Químicos e Bioquímicos/EQ, Universidade Federal do Rio de Janeiro / Programa de Engenharia Química/COPPE, Universidade Federal do Rio de Janeiro)
Programa de Engenharia Química/COPPE, Universidade Federal do Rio de Janeiro Abstract The well drilling process requires constant monitoring to ensure that the properties of the drilling fluids remain within acceptable ranges for safe and effective operation of the well drilling process. The present work developed a principal component analysis (PCA)- based methodology for diagnosing anomalies in drilling fluids, and detecting and identifying abnormal drilling fluid properties during well drilling operations. The main novelty of the present work regards the application of multivariate techniques for diagnosing anomalies (faults) in drilling fluids, increasing the literature on fault diagnosis techniques applied to the petroleum industry, and producing a promising methodology for field applications. The proposed technique was implemented and validated in a pilot drilling fluid production unit through continuous online monitoring of the conductivity, density, and apparent viscosity of drilling fluids. Model training was carried out with data collected during assisted normal operation, allowing detection of abnormal conditions with less than 1% of false positives and less than 0.5% of false negatives. Additionally, the proposed methodology also allowed the correct diagnosis of the observed faults. The results indicated that PCA-based approaches can be used for the online monitoring of drilling fluid properties and fault diagnosis in real well drilling operations. Introduction The oilwell drilling operation is a complex process that demands constant supervision. Process variables must be controlled to ensure that the components of the operation function correctly, with specific corrections being made to operating conditions whenever necessary (Lyons and Plisga 2004). The increasing complexity of the drilling process has motivated the development of new technologies, with emphasis on data analyses and approaches based on artificial intelligence (Li et al. 2021; Kavulya et al. 2012; Mohammadpoor and Torabi 2020). Drilling fluid must perform many other important functions during well drilling (Bourgoyne et al. 1991), such as preventing rock formation fluids from invading the well, keeping the newly drilled hole open until the steel casing can be cemented into the well, and lubricating and cooling the bit while it is running. Drilling fluids are classified in different ways in the literature.
- Asia (1.00)
- Europe (0.93)
- North America > United States > Texas (0.46)
- (2 more...)
- North America > United States > Kansas > Lyons Field (0.93)
- North America > United States > Arkansas > Magazine Field (0.89)
- Europe > United Kingdom > England > London Basin (0.89)
- Information Technology > Artificial Intelligence > Machine Learning > Neural Networks (0.93)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Diagnosis (0.88)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Expert Systems (0.70)
Abstract Contingent Resources do not form part of Reserves reporting requirement for companies listed on most stock exchanges, but they are often provided in the information companies share with investors and the public in the form of media releases. Contingent Resources are, according to the Petroleum Resources Management System (PRMS) "quantities of petroleum estimated to be potentially recoverable from known accumulations but are not yet considered mature enough for commercial development due to one or more contingencies" (PRMS, 2018). The sub-classes in the Contingent Resources classification are: Development Pending Development on Hold Development Unclarified Development Not Viable These sub-classes are not equal in their chance of development and portray a vast range in the "maturity" of a project. From projects that are about to be classified as Reserves, to projects that are unlikely to be developed. The choice of sub-class can have an impact on a company's resource base and it is important to be realistic about the sub-class in which a Contingent Resource belongs. When companies decide to report their Contingent Resources, all of the above sub-classes of the Contingent Resources class are labelled with the letter "C". A requirement to mention the specific sub-class of a Contingent Resource is not always present. This means that without further background information, they all look alike and can appear to have equal maturity and equal chance of development. The PRMS is a system with definitions and guidelines. It is not a set of rules. Thus, terms in the PRMS require interpretation. What do reserves evaluators do to ensure that Contingent Resources are classified and reported consistently? Are investors sufficiently informed when they deal with Contingent Resources, or would it be useful to have a requirement for reporting the respective sub-class? This paper attempts to highlight some of the intricacies of the Contingent Resources class.
- North America > United States (0.28)
- Asia (0.28)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Rule-Based Reasoning (0.34)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Expert Systems (0.34)
ABSTRACT: Metro stations are complex underground infrastructure encompassing wide-span caverns. Their interactions with tunnels, adits, and utilities in vicinity can have impacts on ground stability. This paper aims to undertake a novel approach leveraging digitalization of design and 3D numerical modelling to better visualize and quantify the ground-structure interaction based on a metro station featuring a trinocular cavern platform. Knowledge-based prediction is important for critical areas of underground construction. By engaging numerical modelling and using appropriate constitutive relationships for both short- and long-term deformation criteria, predictions into the ground movement and responses of ground support can be made to offer insights into the adequacy of the support system and propose monitoring scheme correspondingly, which forms an essential part of structural health management for the underground infrastructure when coupled with data-driven analytics. INTRODUCTION Knowledge-based prediction is important for critical areas of underground construction and long-term maintenance. By engaging numerical modelling and using appropriate constitutive relationships for both short- and long-term deformation criteria, predictions into the ground movement and responses of ground support can be made to offer insights into the adequacy of the support system and propose monitoring scheme correspondingly (Huang et al., 2021). With the advance in information communication technologies, a rising number of sensors and gauges are deployed to facilitate monitoring. Numerical modelling has been engaged to investigate long-term, time-dependent deformation of tunnels (Barla et al., 2012) and underground infrastructure such as large-span caverns as part of hydropower facilities (Lee et al., 2019; Yang et al., 2014). For rock mass, literature recorded many studies using empirical approaches for identifying squeezing behavior (Aydan et al., 1996), which is described as large time-dependent convergence during tunnel excavation, as relationships are established based on rock mass quality and excavation conditions and tunnel geometry, e.g., overburden and tunnel span. Constitutive modelling of rock mass creep largely uses elasto-visco-plastic models based on general theory, particularly the overstress theory of (Perzyna, 1966) have been implemented in numerical modelling (Pellet et al., 2009), and are used to calibrate the lab-based behavior and validate parameters as well as simulating the long-term mechanical behaviors of underground construction. The comprehensive general-theory based models typically require high parameter inputs with the parameter determination process considered very time-consuming and cumbersome. Whilst the empirical and rheological creep models implemented in numerical modelling codes as built-in models require relatively less parameters and were verified for adequacy of mimicking the mechanical behaviors of rock under static load for a prolonged time and are thus easier to implement for preliminary analysis of long-term stability and support design.
- Geology > Geological Subdiscipline > Geomechanics (0.49)
- Geology > Rock Type (0.49)
- Information Technology > Knowledge Management > Knowledge Engineering (0.70)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Expert Systems (0.55)
The US Department of the Interior announced measures to enhance worker safety and ensure offshore oil and gas operations on the Outer Continental Shelf are conducted with the utmost safety and oversight standards. The final well control rule from the Bureau of Safety and Environmental Enforcement (BSEE) builds upon the historic regulatory reforms implemented by the department in the aftermath of the 2010 Deepwater Horizon explosion and resulting oil spill that killed 11 people and caused billions of dollars in environmental damage and economic loss to coastal communities. "The Biden/Harris administration is committed to the highest standards of worker safety and environmental protections," said Secretary Deb Haaland. "These improvements are necessary to ensure offshore operations, especially those related to well integrity and blowout prevention, are based on the best available, sound science. As our nation transitions to a clean energy economy, we will continue strengthening and modernizing offshore energy standards and oversight."
- Government > Regional Government > North America Government > United States Government (1.00)
- Energy > Oil & Gas > Upstream (1.00)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Rule-Based Reasoning (0.75)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Expert Systems (0.64)
This study utilized a global oil and gas knowledge base comprising over 1,700 proven conventional hydrocarbon accumulations (C&C Reservoirs, 2023). The containment attributes, including trap, seal, reservoir, fluid, and temperature for all these known hydrocarbon accumulations, provided the foundation for the formulation of the Seal Strength Index (SSI). The influence of each of these attributes on containment capacity of a proven hydrocarbon accumulation is analyzed. A quantitative approach using SSI and global analogues in evaluating the containment of hydrocarbon in a play or prospect has been established (Wu et al, 2022).
- Africa (0.97)
- North America > United States > Gulf of Mexico > Central GOM (0.30)
- Geology > Structural Geology > Fault (0.48)
- Geology > Petroleum Play Type > Conventional Play > Stratigraphic Play (0.36)
- Geology > Geological Subdiscipline > Stratigraphy (0.34)
- North America > United States > Gulf of Mexico > Central GOM > East Gulf Coast Tertiary Basin > Mississippi Canyon > Block 851 > Mars Field (0.99)
- North America > United States > Gulf of Mexico > Central GOM > East Gulf Coast Tertiary Basin > Mississippi Canyon > Block 850 > Mars Field (0.99)
- North America > United States > Gulf of Mexico > Central GOM > East Gulf Coast Tertiary Basin > Mississippi Canyon > Block 808 > Mars Field (0.99)
- (10 more...)
ABSTRACT Unplanned roof falls in tunnel intersections are commonly associated with casualties, accidents, economical loss and therefore pose a serious threat for coal mine profitability. Despite the technological advance in ground control, fall of ground (FoG) in tunnel intersection often occur. Subjective data and information are inevitably used in the design of coal mine development excavations which contribute to unplanned ground instabilities. These aspects are not considered in conventional design tools. Hence, the aim of this paper is to propose expert systems capable of adequately quantify the relationship between the unplanned roof falls, support methods and the geology. Fuzzy inference system (FIS) which uses fuzzy logic to discern relationships between data is implemented in this study. FoG in tunnel intersection data were compiled from several coal mines. Expert knowledge approach was used to construct the FIS. The system parameters include the FoG size, geology, types of support and groundwater conditions. Overall, the results indicated quite good agreement with the field observations. The correct classification rate varied between 67 and 80%. It is concluded that FIS could be used to assess FoG in tunnel intersection in coal mines. INTRODUCTION Despite greater awareness of the environmental implication of the coal business, there is still a considerable demand for it globally. The industrial development in China, India, and Southeast Asia countries were recently reported to be the main drivers of the 0.7% rise in worldwide coal consumption (Zhironkin and Cehlár, 2021). Meanwhile, there are a wide ranges of safety issues associated with coal mining. One of them is fall of ground (FoG) which is a generic term used to describe loose rocks falling from excavation faces (roofs and side walls). Unplanned FoG in roadway intersections in coal mines present a serious threat to mine safety, as intersections are by far the most common area for FoG to happen (Mueller, 2010; Mark and Gauna, 2017). It can occur in many geological and operational situations; even with primary and secondary support structures. There are several reasons for it to happen which include over-spanned intersections, non-sufficient support, excess horizontal stress, and weak geology with low strength of rock intersection.
- Asia (1.00)
- North America > United States (0.68)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Uncertainty > Fuzzy Logic (1.00)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Expert Systems (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Performance Analysis > Accuracy (1.00)
A Platform Floating Algorithm Based on Fuzzy PID with Nonlinear Feedback
Hong, Haochen (School of Naval Architecture and Ocean Engineering, Huazhong University of Science and Technology) | Yang, Shunqi (School of Naval Architecture and Ocean Engineering, Huazhong University of Science and Technology / Kunming Shipborne Equipment Research & Test Center) | Shi, Guannan (School of Naval Architecture and Ocean Engineering, Huazhong University of Science and Technology) | Liu, Gang (Wuhan Second Ship Design and Research Institute) | Xu, Guohua (School of Naval Architecture and Ocean Engineering, Huazhong University of Science and Technology)
ABSTRACT This paper proposes a new algorithm for the floating of the underwater platform. To eliminate the effect of the disturbance and improve the safety of the floating process, a fuzzy PID algorithm with a nonlinear feedback technique is used to control the underwater platform. Compared with the traditional PID algorithm and fuzzy PID algorithm, the proposed algorithm can keep the underwater platform within a smaller angle. In addition, the output of the winches is more smooth compared with the output of the traditional algorithm. The proposed algorithm is more in line with the requirements of the underwater platform floating algorithm. INTRODUCTION With the rapid development of modern science and technology, the field of ship and ocean engineering has played an irreplaceable role in all aspects of modern society. With the increasing depletion of land resources, ocean exploration, as an important subject, has been growing steadily in the past two decades(Fossen, 2011). Among them, the underwater tension leg platform is a very important part. Because the underwater tension leg platform can enter the water while maintaining a large system, it can complete some large-scale experiments and projects. The underwater tension leg platform has a very broad application prospect in resource exploration and crude oil collection. However, the underwater tension leg platform system is huge and involves many systems, so it is difficult to establish a high-precision mathematical model. In addition, in the complex underwater environment, many disturbances need to be overcome in the control process, and the problems of actuator failure and communication failure need to be overcome in the control process. Therefore, the control core of the underwater tension leg platform is still facing huge challenges. n recent years, there are many types of research on the control methods of underwater tension leg platforms. Xia et.al (2018) designed a three-layer sliding mode controller using the uncertainty prediction ability of a neural network and proposed a winch synchronization control strategy. The algorithm makes the motion of TLP more stable and has a stronger anti-interference ability; Zhang and Yang (2016) used the fuzzy PID control method to control the underwater platform, which has a good control effect in angle control and floating control of the platform; Zhang et al. (2014) proposed an observer-based optimal error tolerance control method for offshore steel jacket platforms, which realized real-time observation of hard to observe state variables and improved the reliability and work efficiency of offshore platforms. Nourisola et al. (2015) designed several types of sliding mode controllers to solve the control problem for an offshore steel jacket platform. The new method is very useful to deal with the nonlinear wave force.
- Information Technology > Artificial Intelligence > Representation & Reasoning > Uncertainty > Fuzzy Logic (1.00)
- Information Technology > Artificial Intelligence > Robots (0.88)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Expert Systems (0.70)
Digital Innovation for Fault Diagnosis and Control of Sensor Variables. Reduction in Operational Time and Cost in Run-In-Hole
Mateus, D. (Baker Hughes, Meta, Colombia) | Ramírez, H. (Baker Hughes, Meta, Colombia) | Cruz, J. (Baker Hughes, Meta, Colombia) | González, C. (Baker Hughes, Meta, Colombia) | Pinilla, J. (Baker Hughes, Meta, Colombia)
Abstract The Oil and gas industry is continuously searching for a digital transformation and evolution to Industry 4.0. It plans to complement it whit cutting-edge technologies that positively impacts the installation processes of electro submersible pumping systems (ESP). A few years ago, there wasn't a system capable of giving continuous and on real time ESP downhole diagnosis during Run in Hole operations under any weather conditions and without interruption, situation that usually generated nonproductive times, by not having the ability to identify an electrical fault at the instant it occurred. To reduce periodic stops of the drilling RIG and not to prolong working hours checking electrical conditions, a digital solution was developed with real-time failure identification warning of any electrical and mechanical condition associated with the system (CABLE-MOTOR-SENSOR) during installation. The system is composed by wireless connection to obtain background sensor data reading, with instantaneous fault indicator and web interface, in addition to downloadable data logging that is compatible without connectivity limits for Smartphones and Laptops. By using the system while installing, the time of exposure of personnel to electrical and mechanical risks is reduced by 75% from 4 hours to 1 hour per service and a considerable saving of 16% was obtained in the total time of completion under normal conditions; in the event of an electrical failure with immediate detection, a saving of up to 40% could be achieved. Considering the volume of operations in the country, an economic benefit of US$ $840,000 per year can be obtained. The development and implementation tests were successful, and it was determined to expand the use of the tool during services to increase customer benefits and reduce non-productive times.
- Information Technology > Communications (1.00)
- Information Technology > Architecture > Real Time Systems (1.00)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Expert Systems (0.40)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Diagnosis (0.40)