Summary Faults are natural for any practical system including photovoltaic (PV) power generation systems that need to be diagnosed accordingly to maximize its efficiency. However, faults can be rectified only if they are diagnosed correctly. Therefore, a credible and reliable diagnosis is required for maintenance of the system. In this paper, independent component analysis (ICA)‐based wireless fault diagnosis technique is proposed for solar PV systems. ICA enables us to communicate without having any prior information about the signals and the wireless channel. The proposed technique can diagnose faults related to electrical failures, climatic impacts, variable irradiance, and irregular cell temperature. Effectiveness of the proposed technique is evaluated through extensive simulations utilizing practical data.

Distance-to-bed boundary technologies have been successfully deployed to geosteer wells horizontal sections for the last 10 years. As the technology is based on the propagation of electromagnetic waves, its preferred operational environment is to be in the resistive reservoir layer, while mapping the adjacent conductive layers. Technology advances have driven the development of a more robust inversion engine, as well as opened the opportunity to acquire and transmit a richer suite of measurements in real-time that are more sensitive to reservoir dip and anisotropy. The evolution has taken the technology from distance-to-bed boundary to multi-layer bed boundary mapping improving the resolution from a simple 3-layer model of up to ~15 ft radially from the wellbore to multi-layer detection in excess of 20 ft. The new inversion engine is also better equipped to resolve for information about the reservoir in the less-favorable environment, such as, conductive target reservoir with resistive adjacent layers.

No abstract

New developments in Machine Learning and Artificial Intelligence-based interpretations are bringing a step change in the integration of multi-physic evaluations and management of reservoirs in real time. But it also requires game-changing digital developments to deliver the larger computing power required and to facilitate their access to multi-disciplinary (and sometime not co-located) team of experts and decision makers. This communication is sharing our experience of a web-based collaborative platform integrating operator's application used to produce realistic geological models and a service company's advance multi-dimension modeling and inversion supporting latest Logging While Drilling formation evaluation workflows. The system is now routinely used in case studies, allowing users to perform pre-job well placement feasibility analysis and post-job model refinement. The technology behind is a modular Web platform that hides all the complexity of the modeling and inversions algorithms. Users can; Upload their data to the application's virtual file system. Visualize 2D and 3D models, Launch modeling jobs for Ultra-Deep Azimuthal Resistivity (UDAR) and conventional formation evaluation measurements and finally monitor the inverted images unfold as the job progresses, all in the web browser. The system enables multiple users to view and edit the shared models and observe and control the same job in a collaborative way. The simulation codes are run on the remote clusters or on the cloud. We will present the application of platform and models for 3D characterization in Norwegian continental shelf wells. The examples illustrate mapping of 2D and 3D structural complexity and how the system is used to update reservoir geomodels. The platform is also used to identify optimal well position; define geosteering strategies in the pre-job planning phase, as well as to evaluate sensitivities, depth of investigation in specific scenarios and to analyze how the structural model uncertainties may be affecting the interpretation. Modeling and inversion are used to assess how structural complexities, lithological changes, oil-water contacts and saturation could be encounter in simulating future production. It is a key for quantitative robust interpretation and geomodels update. The platform allows fast deployment of latest research modeling and inversion prototypes. We finally present the latest results of full 3D modeling and various flavors of 2D imaging inversion results from multiple wells, visualized in the browser using a 3D viewer. The new digital solution improves understanding of 3D reservoir structure and fluid distribution around the wellbore.