19 April 2026 · Operations

LNG cargo simulation tools — BOG evolution, weathering, and rollover risk

Boil-off gas, weathering, and rollover are almost always treated as separate problems. In practice they are connected — the same compositional evolution drives all three. A simulator that combines them changes how you approach cargo and storage planning.

Boil-off gas, weathering, and rollover dominate LNG tank management from first fill through long-term storage to sendout. BOG is typically framed as a cargo management and commercial problem, weathering as a custody transfer concern, and rollover as an emergency scenario that belongs in the HAZOP. The simulator package combines all three into a single tool, because they share a common mechanism: the same compositional evolution that drives the Day 10 BOG rate drop is the mechanism that gradually shifts heating value out of specification and can, if density is tracking in the wrong direction, set up a rollover.

The BOG module

The physics engine runs on the Peng-Robinson equation of state with standard van der Waals mixing rules and published binary interaction parameters for the main LNG component pairs — methane-ethane, methane-propane, and so on. Heat ingress is modelled as linear with temperature differential, consistent with standard tank design practice. Tank pressure feeds into the saturation vapour pressure and flash calculation, so BOG rate responds correctly to small pressure perturbations.

The primary display is daily boil-off rate in %/day — the number the industry actually uses — alongside U-value inputs. Compositional evolution is tracked through each time step, which means you can watch the cargo change character over time. Nitrogen, with its boiling point of −196°C against methane's −162°C, preferentially flashes off in the early days. Within the first week or so, depending on initial nitrogen content, the nitrogen fraction is substantially depleted and the BOG rate drops as the remaining cargo shifts toward predominantly methane. The simulator makes this visible rather than burying it in a single aggregate BOR figure.

The UI displays mol% throughout, with explicit units on Wobbe Index, Gross Calorific Value, and density — practical outputs for anyone using the tool in a commercial or custody transfer context.

The weathering module

Weathering is what happens to LNG that sits in storage for weeks or months: the lighter fractions — nitrogen first, then methane to a lesser degree — boil off preferentially, leaving a progressively richer cargo. The liquid gets heavier, its heating value climbs, and eventually the regasified product may fall outside the Wobbe Index or GCV band specified in the gas network entry requirements.

This matters most in three scenarios: long-term strategic reserves held for seasonal or emergency drawdown, FSRU heel periods during extended off-hire or between cargoes, and import terminals receiving spot cargoes of varying provenance that are blended or stored before sendout.

The weathering module tracks GCV and Wobbe Index continuously as a function of compositional evolution. Both are calculated at each time step using the current vapour-phase and liquid-phase composition from the PR EOS flash. The gas network specification limits are user-defined inputs — typically the grid operator's entry point requirements — and the tool flags the projected time to breach on both parameters independently, since GCV and Wobbe do not necessarily go off-spec simultaneously.

For a typical rich LNG cargo starting near the upper end of the Wobbe band, the weathering trajectory moves further into non-compliance over time as methane boils off and the C2+ fraction concentrates. For a lean cargo with high nitrogen content, the early weathering phase can actually improve compliance as nitrogen strips out, before the subsequent methane-dominant phase drives GCV upward. The simulator plots both trajectories — within spec, marginal, and definitively off-spec for the intended receiving network.

The rollover module

Rollover occurs when a tank develops two stable liquid layers of different density: a denser lower layer and a lighter upper layer. The two layers exchange heat with the surroundings independently, and over time the lower layer warms, its density falls, and — if the layers reach equal density before the operator intervenes — they overturn suddenly, releasing the accumulated superheat as a rapid, uncontrolled BOG surge that can exceed the tank's pressure relief capacity.

The trigger is almost always a filling operation where the new cargo has different composition or temperature from the heel. Top-filling into a tank with a dense heel, or bottom-filling into a tank with light, warm LNG already present, can both produce a stable stratified configuration. The risk is not the stratification itself — it's the stability. Layers that mix slowly are safe; layers that stabilise and resist mixing are the problem.

The rollover module tracks two-layer density evolution over time. Inputs are the density and temperature of each layer, heat ingress coefficients for the tank wall and the interface, and the filling scenario that created the stratification. The model integrates forward in time, flagging when the density difference between layers falls below a user-defined threshold — the point at which the stratification becomes unstable and layer inversion can occur spontaneously. It also outputs the projected BOG surge rate at rollover, which you can compare against the tank's vapour handling capacity to assess whether the scenario is within design limits.

Why the three modules belong together

The connection to the weathering module is direct: compositional ageing in the lower layer changes its density independently of temperature. A layer that starts out stable can reach the critical density crossover faster than a temperature-only estimate would predict, because the composition is shifting at the same time. Running the weathering and rollover modules together gives a more complete picture of how a stratified tank evolves over a long storage period.

All three tools address the gap between the process simulator and the engineering judgement call. A HAZOP will identify rollover as a credible scenario and specify mixing pumps as safeguards, but it won't tell you how quickly a given filling scenario reaches critical density crossover or what the BOG surge looks like in quantitative terms. A custody transfer protocol will define the Wobbe and GCV limits, but it won't tell you how many days of storage remain before a held cargo goes off-spec. The simulator package puts numbers on all of these questions, with enough physical fidelity to be useful in an advisory or commercial context.