Risk Assessment Framework for Structural Failures of Polar Ship Under Ice Loads

In ice-covered regions, the upward–downward ice-breaking cones are applied widely to reduce the ice load on the vertical jacket offshore platforms. The ice load is affected by the diameter of cone at the water level. In particular, the sea ice can impact on the vertical portion of the platform which is beyond the covered area of the upward–downward cone under the extreme low or high water level. In this study, the discrete element method (DEM) with the bond and failure model is adopted to simulate the breaking process of sea ice acting on conical structure and vertical pile considering the changing of water level. The ice load and ice failure mode simulated with the DEM is compared well with the field data measured in the Bohai Sea. The DEM results indicate that the bending failure of ice cover occurs when it acts on the upward or downward cone. For both of the upward and downward cones, the ice load increases with the increase in the cone diameter at water level, but the ice load on upward cone is larger than that on downward cone. Meanwhile, the breaking length of ice cover on upward cone is smaller than that on the downward cone. The ice load and the breaking length of ice cover are analyzed based on the DEM results and the field observation in the Bohai Sea. When the ice cover impacts on the junction of upward–downward cone, the ice cover can be broken by a local crushing failure, but it has little effect on the maximum of the ice load. The interaction between ice cover and vertical pile is also simulated with DEM under the extreme low or high water levels. The sea ice breaks with crushing failure mode, and the ice load is obviously larger than that of conical structure. The ice loads and failure modes of ice cover are compared when the ice cover impacts on the vertical pile, upward cone, downward cone and their junction portion considering the influence of water level.

Eliminating the influence of measuring point failure in ice load identification of polar ship structures

Influence of propeller on brash ice loads and pressure fluctuation for a reversing polar ship

Ice loads on a ship hull affect the safety of the hull structure and the ship maneuvering performance in ice-covered regions. A discrete element method (DEM) is used to simulate the interaction between drifting ice floes and a moving ship. The pancake ice floes are modelled with three-dimensional (3-D) dilated disk elements considering the buoyancy, drag force and additional mass induced by the current. The ship hull is modelled with 3D disks with overlaps. Ice loads on the ship hull are determined through the contact detection between ice floe element and ship hull element and the contact force calculation. The influences of different ice conditions (current velocities and directions, ice thicknesses, concentrations and ice floe sizes) and ship speeds are also examined on the dynamic ice force. The simulated results are compared qualitatively well with the existing field data and other numerical results. This work can be helpful in the ship structure design and the navigation security in ice-covered fields.

The determination of ice loads on polar ships and offshore structures is of great significance for ice-resistant design, safe operation, and structural integrity management in ice-infested waters. The physical model testing carried out in ice tank/basin is usually an important technical approach to evaluate the ice loads, however, the high cost and time consumption make it difficult to perform multiple repetitions or large number of trials for this purpose. Recently, the rapid development of high-performance computation techniques provides a usable alternative where the numerical methods represented by the discrete element method (DEM) have made remarkable contributions to the ice load predictions. On basis of DEM simulation validated by physical model testing, numerical ice tank can be developed as an effective supplement to its counterpart. In this paper, such an example of numerical ice tank adopting GPU computational mechanism and DEM modelling algorithm was established with respect to the small ice model basin of China Ship Scientific Research Center (CSSRC-SIMB). The numerical ice tank was calibrated and further optimized with physical model tests on typical structures of vertical cylinder and inclined flat plate in level ice sheets by making agreements of both globe value and time history of the ice loads. Then it was practiced for modelling the tests of Wass bow advancing in level ice performed in SIMB separately. It is demonstrated by the comparisons of ice failure details and ice loads that the numerical ice tank can precisely simulate the ice-structure interactions and determine the ice loads under the same initial conditions of physical model testing. In the end, the advantages as well as the challenges of the numerical ice tank are discussed.

Existing standards and codes do not comprehensively address and provide all necessary guidance and requirements for the design of Arctic offshore structures. Reliably assessing ice loads on offshore structures remains a challenge for industry, especially as "new" Arctic platform concepts are proposed for deployment in lighter ice operations, e.g. Arctic SEDU (Self- Elevating Drilling Units) and Arctic CSDU (Column-Stabilized Drilling Units). As a pragmatic solution, a comprehensive approach including relevant field measurements, physical model tests and numerical simulations is usually adopted in assessing the ice loads for a particular design. Field measurement results are limited in number and there is often uncertainty concerning actual ice conditions and load measurement techniques. Furthermore several data sets are constrained with proprietary restrictions. In particular no direct field data is available for the example "new" structures (SEDUs and CSDUs). Physical model tests always present challenges due to scale issues, ice property calibration, measurement uncertainties and high costs. Therefore, numerical simulations using the validated DEM tool based on the related field/model test data are anticipated to provide supplementary information for standards/rule-based designs. ABS has expended efforts to develop practical and advanced tools to assess the ice loads on offshore structures for several years. One promising numerical approach, a graphic processing unit (GPU) based Discrete Element Method (DEM) model, processes the computations in parallel and solves the DEM model with millions of particles for complicated ice-structure interaction problems, e.g. ice simultaneously loading on multiple legs and ice loading on a large CSDU. The paper presents details of the developing ABS GPU-DEM tool, status of the verification program, plans for the current and the future developments and applications. Included are brief descriptions of background technologies, an approach to derive the DEM model bonding strength inputs, and validation studies of ice breakage simulations based on the Bohai Bay Jacket ice data. The ice load simulations for fixed and floating structures, i.e. jack-up legs and the Kulluk floating drilling platform, are also shown to demonstrate the tool's capability and feasibility for Arctic offshore structure design. The interactions of ice and fixed/floating structures were analyzed, which provides useful references for future ice load modelling and offshore structure design.

In recent years, the melting of Arctic ice has intensified due to the continuous increase in the global average temperature, which has made it possible to open Arctic channels. Predicting ice resistance during ice breaking is a crucial issue for navigating polar ships in ice-covered waters. However, studies on the dynamic ice resistance hydrodynamics and ship motion response of subsequent polar ships in channels after ice breaks are rare. This paper employs CFD-DEM technology to investigate the impact of a channel on a polar ship after icebreaking. The contact of discrete particles is simulated with a variant of the spring-damper model. The fifth-order Stokes wave is introduced to simulate wave action based on nonlinear wave theory. The simulation uses overlapping grid technology to model the ship’s motion response under the combined effects of nonlinear waves, currents, and brash ice. The fluid control equation is discretized using the finite volume method, while the volume of fluid method is used to capture the free surface. Finally, the grid independence is verified and compared with that of the direct sailing resistance test of the bare hull. The numerical simulations of the ice impact force, nonlinear wave action, and ship-ship coupling state were found to be in good agreement. The coupling of CFD-DEM methods can enhance the reliability and efficiency of simulating collisions between structures and brash ice. The model developed in this study can be used to investigate the coupling between dual structures and brash ice collisions.

The icebreaking research vessel, ARAON had her second ice trial in the Arctic Sea from 16th July to 12th August 2010. During the voyage, the local ice loads acting on the bow of port side were measured from 14 strain gauges. The measurements were also carried out in ice waters with various ice concentration ratio as well as the icebreaking performance tests. In this study, the ice loads measured during the 'general' operation in ice waters were analyzed. As a first step, the relationship between the location of strain gauges and the ice loads were investigated, and then the possibility for observation of higher ice loads was estimated based on the probability density function. The relationship between the ship speed and the ice load was also investigated. 718 peak stresses data higher than 20 MPa obtained from strain gauges array attached in longitudinally and vertically was analyzed. In general, the ice load increases as the ship speed increases in the low ship speed range, and ice load decreases as the ship speed is greater than a certain speed.

Purpose – In oil/gas exploitations of ice-covered cold regions, conical offshore structures are designed to reduce ice force and to avoid the ice-induced intense vibrations of vertical structures. The purpose of this paper is to investigate the interaction between ice cover and conical offshore structures, the discrete element method (DEM) is introduced to determine the dynamic ice loads under different structure parameters and ice conditions. Design/methodology/approach – The ice cover is dispersed into a series of bonded spherical elements with the parallel bonding model. The interaction between ice cover and conical offshore structure is obtained based on the DEM simulation. The influence of ice velocity on ice load is compared well with the experimental data of Hamburg Ship Model Basin. Moreover, the ice load on a conical platform in the Bohai Sea is also simulated. The ice loads on its upward and downward ice-breaking cones are compared. Findings – The DEM can be used well to simulate the ice loads o...

Dynamic performance optimization of an arctic semi-submersible production system

The ice loads on ship and offshore platform structures is the key factor in structure design for cold regions. The discrete element method (DEM) is an important approach to determine the ice load on structures. According to the Minkowski sum theory, the dilated polyhedra based DEM is employed to simulate the interaction between sea ice and ship and offshore platform structures in this paper. In the dilated polyhedra based DEM, the enveloped function of the dilated polyhedron is generated to establish the fast contact detection algorithm based on the optimization model. Meanwhile, the bond-break model between elements is established by considering the stiffness softening process between bonded elements. Accordingly, the high-performance algorithm based on CPU-GPU cooperative-heterogeneous environment is developed. The ISO standard is employed to validate the ice load determined by the dilated polyhedra based DEM for better engineering applications of the interaction between sea ice and marine structures. The ice load on ship hull is calculated by the proposed method while the line load distribution on ship hull is studied. The ice resistance of ship hull is compared with the result by Lindqvist empirical formula to validate the accuracy of DEM simulations. The interaction between level ice and multi-leg platform is simulated while the ice load on each leg is analyzed. For the ice management in broken ice regions, the ice load on ship and offshore structures is simulated when the ship navigates around the offshore platform in circle. The proposed method can be effectively applied in the analysis of ice load on marine structures, and can provide a scientific approach for the design and safety operation of ship and offshore structures.

Scenario-based optimization design of icebreaking bow for polar navigation

The determination of ice loads on polar vessels and offshore structures is important for ice-resistant design, safe operation, and management of structural integrity in ice-infested waters. Physical model testing carried out in an ice tank/basin is usually an important technical approach for evaluating the ice loads. However, the high cost and time consumption make it difficult to perform multiple repetitions or numerous trials. Recently, the rapid development of high-performance computation techniques provides a usable alternative where the numerical methods represented by the discrete element method (DEM) have made remarkable contributions to the ice load predictions. Based on DEM simulations validated by physical model tests, numerical ice tanks can be developed as an effective complement to their counterparts. In this paper, a numerical ice tank based on 3D spherical DEM was established with respect to the small ice model basin of China Ship Scientific Research Center (CSSRC-SIMB). Based on spherical DEM with parallel bond model, the model tests of typical structures (vertical cylinder and inclined plate) in level ice sheets were established in the numerical ice tank, and the ice–structure interaction process under the same initial conditions was simulated. The accuracy of the simulations is verified by comparing the simulated ice loads with the measured ice loads from the model tests in the CSSRC-SIMB. Furthermore, the application of the numerical ice tank was extended to simulate the navigation of a Wass bow in level ice and broken ice conditions. The value of the break resistance of the Wass bow in level ice was evaluated, and the numerical ice tank produced results that were found to be consistent with those obtained from Lindqvist’s formula. The statistical properties of the bow load for different broken ice fields with the same initial physical conditions are analyzed by performing a repeatability test on the broken ice fields.

Ice loads, due to large absolute values and diversity, are most important for the calculation of Arctic marine and river structures. The cost of the structure being designed may change several times with a change in the design ice load; therefore, determining the ice load of a given probability is an important task. Some ice formations can significantly change the value of the ice load, but they are not present in the normative documents and standards. Ice collars can be described as thicker ice around heat-conducting structures, which can lead to an increase in loads. However, there is not any method for calculation of ice collar influence on the loads present in the current documents. Ice collar is the reason for thicker ice at the structure-ice contact surface, which lead to an increase in the contact area and can lead to a change in the mechanism of ice failure. In some cases, the presence of an ice collar leads to an increase in the horizontal ice load from the level ice on the structure. The method of discrete elements is considered to solve the problem of the influence of the ice collar on the ice loads. This method allows a time-dependent description of the entire process of destruction of the ice collar. Two-dimension settings are used in vertical and horizontal planes. The relative ice load depending on the thickness of the ice and freezing-degree-days are obtained.

Apart from breaking level ice, polar ships can interact with broken ice in various scenarios. In recent years, computational simulation models have increasingly been used for the evaluation of ship operability under broken ice conditions, presenting some challenging issues. This paper reviews existing simulation methods used to estimate ship performance and ice loads for ships advancing continuously in broken ice fields. Models for different types of broken ice, including ice floes, ice ridges, brash ice, and sliding ice pieces, are reviewed separately. A ship’s response in broken ice is divided into two categories: resistance, which relates to the overall ship performance, and local loads, which relates to structural safety. This review shows that most existing models are proposed for unbreakable ice particles, which are only applicable to broken ice of small size; most models treat fluid flow with extensive simplification, which does not reflect the influence of a ship’s wake or bow waves, and most models are aimed at resistance estimation, adopting elastic or viscoelastic contact models which do not include ice crushing. As for future work, it is suggested that more effort should be assigned to simulating a ship’s interaction with ice ridges and sliding ice pieces, the modelling of breakable ice floes, and the coupling of the Discrete Element Method (DEM) and Computational Fluid Dynamics (CFD). More attention to the local ice load estimation is also encouraged.