Temperature Field Prediction of a Multilayered Composite Pipeline Based on the Particle Filter Method

Thermal management of subsea oil production systems in deep water environments is one of the main issues for petroleum exploitation operations. Thermal monitoring is crucial to avoid and control the formation of solid deposits, which in adverse operating conditions can result in blockages inside the production systems and consequently incur large financial losses. This paper aims to demonstrate the robustness of a Bayesian approach for accurate estimation of the produced fluid temperature field in a typical multilayered composite pipeline. The physical problem consists of a pipeline represented by a circular domain filled by a stagnant fluid (petroleum) with temperature dependent thermal properties, which is bounded by a multilayered composite pipe wall. The mathematical model governing the heat conduction problem in the multilayered wall and in the stagnant fluid was solved with the finite volume method. The Particle Filter method was used for the solution of the inverse transient problem involving the prediction of the temperature field in the medium, from limited temperature data available at one single location in the pipeline composite wall. The aim of this method is to represent the required posterior density function by a set of random samples with associated weights, and to compute the estimates based on these samples and weights. Results are presented in this paper by taking into account uncertainties in the state evolution and measurement models. Simulated temperature data is used in the inverse analysis for typical conditions observed during production shutdown periods.