Scales, D., Hubbard, R. & Merrick, C. (2018) “Risk Based, Statistical Approach to Assessment of Corrosion Anomalies in Pipelines”, Offshore Technology Conference 2018, Houston, (OTC)

In-line inspection of subsea pipelines is a common method to detect anomalies within carbon steel pipelines however the cost of an intelligent pigging campaign is sometimes not economically viable, particularly for situations requiring subsea to subsea pigging. DNVGL-ST-F101 is a risk based offshore standard which requires that the probability of failure be lower than pre-determined levels depending on the safety class of the Pipeline system. As a consequence a risk based assessment methodology has been developed and utilized to ensure code compliance.

This paper presents a methodology and a supporting case study for the use of an externally mounted inspection tool to undertake spot checks of the pipe wall to determine an overall statistical corrosion model of the pipeline in lieu of more costly in-line deterministic inspection methods. This inspection was required to support a pipeline’s commissioning process after wet parking of the pipeline for an extended period.

The methodology presented utilizes the results from a spot inspection tool that can report wall thickness for discrete elements of the pipe wall (for example 1 mm2 or 1 cm2) with each scan encompassing a short length of the pipeline (for example a 1 m length) resulting in a large number of data points. The data points are samples from the overall pipe population and as such can be used to infer the overall pipe wall element population’s statistical parameters. To provide confidence that the data points are representative of the overall population it is important to understand the credible corrosion mechanisms within the pipeline and ensure that sampling locations are chosen accordingly to capture areas in which high corrosion may occur. Sub-populations may also be identified based on key locations (e.g. girth welds) and assessed separately to provide a refined assessment of the overall corrosion mechanisms. Statistical boot-strapping is performed on the data points to provide confidence bands around the sampled data with the most conservative extreme probability curve being utilized in the final assessment.

The case study provides details on potential pipe wall sub-populations and the corresponding corrosion mechanisms for a wet parked pipeline within PTTEP’s Montara field offshore north-western Australia based on data from a spot inspection campaign that was undertaken. The Montara Production Flowline 1 was wet parked for operational reasons during commissioning of the field in 2012 and remained isolated while corrective work was undertaken on Oil Separation train A. The purpose of the analytical work detailed within this paper was to facilitate a fitness-for-purpose assessment of the flowline, integrating the inspection results gained from externally applied ART provided by HalfWave, the external inspection history and the as-built commissioning records from 2012. By returning the flowline to service, a newly drilled and completed platform well was able to be flowed through a quarantined separation train without the clean-up fluids interfering with a working production train. Commissioning of this flowline restored the flexibility for production operations contemplated in the facility design and will improve the availability of production going forward.

Application of the statistical assessment methodology to the pipeline inspection data detailed within the case study resulted in a cost reduction of up to AU$15M compared to in-line inspection.