The paper deals with the problem of controlling the state of industrial devices according to the readings of their sensors. The current methods rely on one approach to feature extraction in which the prediction occurs. We proposed a technique to build a scalable model that combines multiple different feature extractor blocks. A new model based on sequential sensor space analysis achieves state-of-the-art results on the C-MAPSS benchmark for equipment remaining useful life estimation. The resulting model performance was validated including the prediction changes with scaling.

Ice management is one of the ways to achieve a more efficient operation of marine ice-resistant platforms in winter, to secure oil offloading operations and, if necessary, to support rescue and evacuation operations. Since 2012, Krylov State Research Centre, St. Petersburg, Russia, runs a training complex for navigational simulation of ship movement in ice conditions, developed with participation of TRANSAS, a global leader in the field of marine navigation simulators. The core functional elements of this training complex are the models of interaction between marine technology (ships and platforms) and ice in given water areas. One of the completed Russian Projects, Prirazlomnaya offshore ice-resistant platform, is taken as a case study to illustrate a stage-by-stage process of working out a tactic of ice management near the platform, including: 1) analysis of ice conditions in the area of Prirazlomnoye oil field, with consideration of environmental variability; 2) development of maneuvering procedures for the ice-breaking vessels to ensure safe oil offload to tankers; 3) polishing these maneuvers by means of a training complex; 4) analysis of the real maneuvers performed near the platform. This approach allowed a deeper understanding of the developed maneuvering layouts for supply vessels, with respect to specific ice conditions and given operational circumstances.

No abstract

From experience, OIRFP Prirazlomnaya (offshore ice-resistant fixed platform) installed in the Pechora Sea was seen to accumulate sustained piles of rubble ice in winter time. These features were formed along the platform walls and rested on the seabed (berm) obstructing operation of supply ships and crew evacuation. Researchers from the Krylov State Research Centre, St. Petersburg, were studying the ice piling processes and behavior under variation of environmental conditions (e.g., changes in ice drift headings) during design as well as service life of the platform. Moreover, the Krylov Centre ice basin was engaged in development of various ice management techniques to reduce the ice buildup as well as to remove rubble ice piles. The results of these investigations, which are presented in the paper, have led to a relatively simple solution for removing the rubble ice piles with a special-purpose shovel (excavator) mounted on the deck of a supply vessel. The paper describes details of the suggested solution and its implementation. The efficiency of this ice pileup management method has been proven in actual operating conditions.

No abstract