We are proud to announce that in 2024, ALFAsim had two new releases, ALFAsim 2024.1 and ALFAsim 2024.2. ALFAsim is a tool developed by ESSS O&G to tackle flow assurance problems such as hydrates, which are particularly common in offshore production wells.   Water, under  certain temperature and pressure conditions, can promote hydrate formation. 

These ALFAsim updates include a new Hydrates Plug-in. This plug-in takes into account a porosity growth model and a new capillary bridge mechanism for hydrate agglomeration.  A powerful tool for gas hydrates transport management, this plug-in also considers a slurry viscosity model and a wall deposition mechanism for more accurate predictions.

There were incorporated new functionalities related to the uncertainty quantification framework, as the new Local and Global Sensitivity Analysis, History Matching Analysis, and Uncertainty Analysis features, designed to create a methodology for decision-making in Flow Assurance based on risk quantification, were also added to ALFAsim.

Major highlights of ALFAsim version 2024.2

• Hydrates Plug-in
  • Formation and agglomeration model based on the evolution of the hydrate crystals porosity, following the Bassani (2020) model.
  • Slurry viscosity model based on Camargo and Palermo (2002) for hydrates transportation management.
  • Models wall deposition using Di Lorenzo (2018) and Wang (2017).
  • New thermodynamic inhibition based on the Hu-Lee-Sum model (Hu et al., 2018).

• Uncertainty Analysis
  • Provides visualization of the credible interval in an ALFAsim’s output variable, a plot that estimates the most probable values of output variables given the probability distribution of input variables (parametric variables).
  • User Defined Probability Distributions
    • User sets the uncertainty associated with an input variable (parametric variable) as the uniform probability distribution
    • Monte Carlo Algorithm is used for propagation
    • Credible interval plot
  • Calibrated Probability Distributions
    • The calibrated probability distribution is calculated/quantified with respect to the historical data input by the user in the Probabilistic History Matching.
    • Credible interval plot

Major highlights of ALFAsim version 2024.1

• Sensitivity Analysis
  • Local Sensitivity Analysis
    • Performs specific perturbations of parameter values, positive and negative
    • Provides a tornado plot

• Global Sensitivity Analysis
  • Evaluates a specific model (ALFAsim) to understand objectively which parameter has the most influence in the simulator outcomes and in which conditions.
  • Monte Carlo method offers parameters combinations.
  • Morris Method, or Elementary Effects Method, Saltelli (2004); Morris (1991): the definition of Quantities of Interest (QoI) is used to calculate the indexes and the range of values each parameter can have.

• History Matching Analysis
  • Deterministic – Optimization
    • An optimum value is found for each parameter used to fit the model/simulator.
    • Local optimization methods are used: Nocedal and Wright (2006), SLSQP (Sequential Least SQuares Programming) method, and Kraft (1988).
    • A global optimization method is used: Differential evolution method, Storn and Price (1997).

• Probabilistic – Bayesian
  • Determines a probability distribution function that represents which  parameter values have the highest probability of making the simulation outcomes values close to the historical field data.
  • Bayes Theorem, Meyer (1970).
  • Markov Chain Monte Carlo, Smith (2013) and McClarren (2018).

We are committed to providing the best possible experience for our users, and your feedback is fundamental in helping us achieve that goal. If you encounter any issues or have suggestions for improvements, contact our experts today.