Reliability and safety assessment of Asubsea rov-based sensor retrieval tool

Abstract

The ROV –based subsea sensor retrieval system is a new technology designed by RoxarFlow Measurement As, for the installation and retrieval of subsea sensors mounted on subsea production equipment (such as Christmas trees or Manifolds). The subsea sensor retrieval system consists of four main components, namely; a permanently installed mechanical interface, sensor/transmitter, Remotely Operated Vehicle‐based sensor retrieval tool, and hydraulic and control system. The retrieval process is performed within a controlled, leak‐proof underwater (subsea) environment.

This master thesis focuses on the reliability and safety assessment of the “sensor retrievaltool” which is a sub‐unit of the entire system function. A qualitative and quantitative reliability and safety assessment approach was carried out in compliance with the DNV RPA203 regulation and the API RP 17N; 2009, standard. The introduction of the DNV RP A203 regulation is based on the fact that the system under consideration is a new technology and certain guidelines are needed to assess the reliability and safety of such kind of technology. While, the API RP 17N; 2009 implementation addresses the issue of safety interms of risk and also establishes technique to manage the reliability/risk assessment processes. Other standards such as the ISO 12100; 2010/ISO14121; 1999, were introduced during the risk assessment approach.

A literature study on the qualification and quantification of the reliability/risk assessment methods in compliance with the stated regulation and standards above reveals that certain methods such as the Failure Mode Effect and Critical Analysis (FMECA) and Preliminary Hazard Analysis (PHA) is suitable for an early design phase reliability /risk qualification assessment in order to identify deviations in the design phase of the retrieval tool. The quantification of the retrieval tool was based on estimating the Mean time to failure(MTTF) of each of the component within the system function. Numerical estimates were made by assuming Weibull distribution for the components since the technology is new with little or no historical data.

Recommendation for further work was highlighted in areas where the writer believes that further clarification may be needed in order to achieve a complete reliability and safety assessment.