Project ECSADES

Environmental Contours for SAfe DEsign of Ships and other marine structures (ECSADES)

The ECSADES project addressed important and pertinent issues related to the safe design and operation of ships and other marine structures. Such structures need to be design according to tolerable risk levels. However, in order to understand the risks, the understanding of the extreme ocean environments and their interaction with fixed and floating structures is critical. In this regard, design contours are useful in describing the joint behavior of various environmental parameters, structural loading and response variables.

Currently, there are different approaches to establishing environmental contours, and they all have certain strengths and weaknesses and different mathematical and statistical properties. One main objective in this project was to further develop one of the novel approaches to estimating environmental contours with clearly understood probabilistic properties. Moreover, a detailed comparison study involving climate descriptions from different important ocean regions and various structural problems were performed in order to better understand the effect of the estimation method on the resulting design and reliability levels. Furthermore, some interesting and challenging special cases were investigated, and methods for contour estimation in non-trivial cases have been explored. In addition, the conditional extremes model for joint extremes have been explored and a marginal non-stationary and conditional extremes model have been elaborated to account for covariate effects in the description of multivariate extremes modelling. 

Project results are thoroughly documented in various deliverables and publications in form of journal articles and conference presentations. In addition, several seminars have been arranged where project results have been disseminated internally within DNV. Moreover, project results have been implemented in two software packages that have been developed within the project and that are made freely available through the project GitHub site (see below). The first is a MATLAB code for the Penalized Piecewise constant conditional extremes model and the second is an R packages for generating various environmental contours from the conditional extremes model and for a particular conditional model. 

Finally, the DNV recommended practice on environmental conditions and environmental loads has been updated with a description of the direct sampling approach to environmental contours which has been further developed within this project. 

The software developed within the project is available here: https://github.com/ECSADES

Some deliverables from the project are available here: 

LIST OF PUBLICATIONS from the project:

Technical DNV reports (may be obtained from DNV library): 

  • D1 - Environmental contours comparison Studies. Erik Vanem. DNV report No. 2017-0163, ECSADES
  • D4 – Environmental Contours for Special Cases. Erik Vanem, Odin Gramstad, Bingjie Guo. DNV report No. 2019-1072, ECSADES

Journal papers: 

Conference proceedings papers

Conference and workshop presentations: 

International training course

  • Statistical modelling of extreme environmental conditions for marine engineering applications.
    Erik Vanem. Lecture at International Offshore Structures Design Course 2019, Porto, Portugal. May 7-10 2019. 

Internal seminars: 

Several internal seminars and presentations have been held at DNV headquarters, to disseminate project results. The presentations from these seminars are made available here: 

Seminar 1: January 26. 2017. 

Seminar 2: March 8 2017. 

  • Mini-course on the use of environmental contours (Ø. Hagen, DNV)
  • Mini-course on the conditional extremes model (P. Jonathan, Shell)
  • Coastal applications of multivariate extreme value techniques (B. Gouldby, HR Wallingford)

Seminar 3: May 21 2019.

Seminar 4: October 15. 2019: ECSADES final seminar. 
See images from the event here.

Project partners

  • DNV Strategic Research & Innovation
  • Shell UK
  • Department of mathematics, University of Oslo
  • With participation also from HR Wallingford