SKIP TO PAGE CONTENT
Search profiles
Profile image for Velimir  Vesselinov
  • Velimir Vesselinov

  • Task Lead for "Decision Support", EES-16
  • Email
  • (505) 665-1458
  • machine learning, blind source separation, feature extraction, data compression, exploratory analysis, unsupervised/supervised/deep machine learning

  • model inversion, parameter estimation, uncertainty, sensitivity and risk analysis, performance assessment  and decision support (MADS; http://mads.lanl.gov)

  • experimental and remediation design

  • simulation of multiphase flow, transport and biogeochemical reactions in porous/fractured media (CHROTRAN; http://chrotran.lanl.gov)

  • analytical and numerical techniques for simulation of flow/transport in saturated/unsaturated, porous/fractured media;

  • development of regional/site scale conceptual and numerical models; evaluation of conceptual model uncertainties; ontological trees;

  • parameter estimation; characterization of heterogeneity of subsurface flow medium; high-resolution stochastic imaging (tomography); stochastic inverse analysis; geostatistical and Monte-Carlo Markov-Chain methods;

  • scale effects in medium properties; Lévy (alpha-stable) distributions; fractal formalism;

  • subsurface fluid dynamics and contaminant transport; well hydraulics; exploration and protection of groundwater resources; design of groundwater-supply systems; capture-zone analyses;

  • decision support, model-based decision making, impact evaluation of uncertainties on decision making; risk assessment;

  • quantification of uncertainty associated with estimates, predictions, and conceptual model elements;

  • model selection; model ranking; Maximum Likelihood Bayesian Averaging (MLBA); Generalized Likelihood Uncertainty Estimation (GLUE);

  • value of information; data-worth analyses; global sensitivity and uncertainty analyses;

  • optimal design of environmental management activities (characterization, data acquisition; remediation; monitoring); experiment design aiming reduction of data gaps and uncertainties; optimal design of monitoring networks; decision trees;

  • General Information Theory (GIT); Fuzzy sets; Rough sets; Bayesian techniques;

  • single- and multi-objective optimization methods; global/local techniques; Levenberg-Marquardt and Particle Swarm methods;

  • model abstraction; model reduction; reduced order modeling;

  • high-performance programming;

  • quantum computing.