The impact of data errors on uncertainty analysis

Gillian Elizabeth Pickup, Michael Andrew Christie, Malcolm Sambridge

    Research output: Contribution to conferencePaper

    Abstract

    Since there is much uncertainty in reservoir modelling, it makes sense to start with coarse-scale models, so that a wide range of scenarios can be assessed rapidly, before focussing on fewer, more detailed models. The simplest model for reservoir analysis is the material balance equation, and this forms a good starting point for uncertainty appraisal. Although there are drawbacks with this method, such as the assumption of pressure equilibration throughout the reservoir (or compartment), there is the advantage that a minimum number of a priori assumptions are made regarding the reservoir volume and drive mechanism. As the first stage in a top-down reservoir evaluation procedure, we have applied stochastic history matching and uncertainty analysis to a material balance problem, using a synthetic reservoir model which had aquifer influx and high rock compressibility. A truth case simulation was run and noise was added to the resulting fluid production and pressure values to generate synthetic data sets. The parameters adjusted were the volume of oil (STOIIP), the initial aquifer size and the rock compressibility. A thorough analysis of the errors was performed, including propagation of errors in the pressure data to determine their effect on the modelled production. The Neighbourhood Approximation (NA) method was used to home in on models with low misfit. Then the posterior probability distributions and their correlations were assessed using a Bayesian approach. Results showed that the shape of the posterior probability distributions (PPDs) depended on the assumed level of the noise. In particular, they indicated that, if the amount of noise is not assessed correctly, the position of the maximum likelihood value may be estimated incorrectly.
    Original languageEnglish
    Pages1-9
    Number of pages9
    Publication statusPublished - Sep 2006
    Event10th European Conference on the Mathematics of Oil Recovery 2006 - Amsterdam, Netherlands
    Duration: 4 Sep 20067 Sep 2006

    Conference

    Conference10th European Conference on the Mathematics of Oil Recovery 2006
    Abbreviated titleECMOR X
    CountryNetherlands
    CityAmsterdam
    Period4/09/067/09/06

    Fingerprint

    uncertainty analysis
    compressibility
    aquifer
    rock
    fluid
    oil
    history
    modeling
    simulation

    Cite this

    Pickup, G. E., Christie, M. A., & Sambridge, M. (2006). The impact of data errors on uncertainty analysis. 1-9. Paper presented at 10th European Conference on the Mathematics of Oil Recovery 2006, Amsterdam, Netherlands.
    Pickup, Gillian Elizabeth ; Christie, Michael Andrew ; Sambridge, Malcolm. / The impact of data errors on uncertainty analysis. Paper presented at 10th European Conference on the Mathematics of Oil Recovery 2006, Amsterdam, Netherlands.9 p.
    @conference{1eb5dd6313eb4b0bbcf750d62bfd694e,
    title = "The impact of data errors on uncertainty analysis",
    abstract = "Since there is much uncertainty in reservoir modelling, it makes sense to start with coarse-scale models, so that a wide range of scenarios can be assessed rapidly, before focussing on fewer, more detailed models. The simplest model for reservoir analysis is the material balance equation, and this forms a good starting point for uncertainty appraisal. Although there are drawbacks with this method, such as the assumption of pressure equilibration throughout the reservoir (or compartment), there is the advantage that a minimum number of a priori assumptions are made regarding the reservoir volume and drive mechanism. As the first stage in a top-down reservoir evaluation procedure, we have applied stochastic history matching and uncertainty analysis to a material balance problem, using a synthetic reservoir model which had aquifer influx and high rock compressibility. A truth case simulation was run and noise was added to the resulting fluid production and pressure values to generate synthetic data sets. The parameters adjusted were the volume of oil (STOIIP), the initial aquifer size and the rock compressibility. A thorough analysis of the errors was performed, including propagation of errors in the pressure data to determine their effect on the modelled production. The Neighbourhood Approximation (NA) method was used to home in on models with low misfit. Then the posterior probability distributions and their correlations were assessed using a Bayesian approach. Results showed that the shape of the posterior probability distributions (PPDs) depended on the assumed level of the noise. In particular, they indicated that, if the amount of noise is not assessed correctly, the position of the maximum likelihood value may be estimated incorrectly.",
    author = "Pickup, {Gillian Elizabeth} and Christie, {Michael Andrew} and Malcolm Sambridge",
    year = "2006",
    month = "9",
    language = "English",
    pages = "1--9",
    note = "10th European Conference on the Mathematics of Oil Recovery 2006, ECMOR X ; Conference date: 04-09-2006 Through 07-09-2006",

    }

    Pickup, GE, Christie, MA & Sambridge, M 2006, 'The impact of data errors on uncertainty analysis', Paper presented at 10th European Conference on the Mathematics of Oil Recovery 2006, Amsterdam, Netherlands, 4/09/06 - 7/09/06 pp. 1-9.

    The impact of data errors on uncertainty analysis. / Pickup, Gillian Elizabeth; Christie, Michael Andrew; Sambridge, Malcolm.

    2006. 1-9 Paper presented at 10th European Conference on the Mathematics of Oil Recovery 2006, Amsterdam, Netherlands.

    Research output: Contribution to conferencePaper

    TY - CONF

    T1 - The impact of data errors on uncertainty analysis

    AU - Pickup, Gillian Elizabeth

    AU - Christie, Michael Andrew

    AU - Sambridge, Malcolm

    PY - 2006/9

    Y1 - 2006/9

    N2 - Since there is much uncertainty in reservoir modelling, it makes sense to start with coarse-scale models, so that a wide range of scenarios can be assessed rapidly, before focussing on fewer, more detailed models. The simplest model for reservoir analysis is the material balance equation, and this forms a good starting point for uncertainty appraisal. Although there are drawbacks with this method, such as the assumption of pressure equilibration throughout the reservoir (or compartment), there is the advantage that a minimum number of a priori assumptions are made regarding the reservoir volume and drive mechanism. As the first stage in a top-down reservoir evaluation procedure, we have applied stochastic history matching and uncertainty analysis to a material balance problem, using a synthetic reservoir model which had aquifer influx and high rock compressibility. A truth case simulation was run and noise was added to the resulting fluid production and pressure values to generate synthetic data sets. The parameters adjusted were the volume of oil (STOIIP), the initial aquifer size and the rock compressibility. A thorough analysis of the errors was performed, including propagation of errors in the pressure data to determine their effect on the modelled production. The Neighbourhood Approximation (NA) method was used to home in on models with low misfit. Then the posterior probability distributions and their correlations were assessed using a Bayesian approach. Results showed that the shape of the posterior probability distributions (PPDs) depended on the assumed level of the noise. In particular, they indicated that, if the amount of noise is not assessed correctly, the position of the maximum likelihood value may be estimated incorrectly.

    AB - Since there is much uncertainty in reservoir modelling, it makes sense to start with coarse-scale models, so that a wide range of scenarios can be assessed rapidly, before focussing on fewer, more detailed models. The simplest model for reservoir analysis is the material balance equation, and this forms a good starting point for uncertainty appraisal. Although there are drawbacks with this method, such as the assumption of pressure equilibration throughout the reservoir (or compartment), there is the advantage that a minimum number of a priori assumptions are made regarding the reservoir volume and drive mechanism. As the first stage in a top-down reservoir evaluation procedure, we have applied stochastic history matching and uncertainty analysis to a material balance problem, using a synthetic reservoir model which had aquifer influx and high rock compressibility. A truth case simulation was run and noise was added to the resulting fluid production and pressure values to generate synthetic data sets. The parameters adjusted were the volume of oil (STOIIP), the initial aquifer size and the rock compressibility. A thorough analysis of the errors was performed, including propagation of errors in the pressure data to determine their effect on the modelled production. The Neighbourhood Approximation (NA) method was used to home in on models with low misfit. Then the posterior probability distributions and their correlations were assessed using a Bayesian approach. Results showed that the shape of the posterior probability distributions (PPDs) depended on the assumed level of the noise. In particular, they indicated that, if the amount of noise is not assessed correctly, the position of the maximum likelihood value may be estimated incorrectly.

    M3 - Paper

    SP - 1

    EP - 9

    ER -

    Pickup GE, Christie MA, Sambridge M. The impact of data errors on uncertainty analysis. 2006. Paper presented at 10th European Conference on the Mathematics of Oil Recovery 2006, Amsterdam, Netherlands.