Reducing the Relative Error Between the Experimental and Numerical Results of a Pipeline Leak Flowrate Using a Lean Six-Sigma Based Approach

Abstract

One of the limitations of pipelines performance and structural integrity assessment is the continuous inspection of possible leaks due to corrosion or other types of failure mechanisms. Efforts to develop new technologies started several decades ago where different inspection techniques were used to enhance pipelines structural integrity. A research project has started with the goal of monitoring the pipeline health and detecting leaks and their location and size once they occur. One of the steps of the research project is to conduct an experimental study mimicking a leaking pipeline and develop correlations between leaks and their location and size. The goal of the experiment is to validate the simulation model by comparing the measured data with the simulation results. If the difference is negligible and the model is verified, the rest of the tests where the leak location and size are varied will be conducted only in the simulation software to avoid additional experimental cost and time. However, after running the first experiment, the recorded data did not match the simulation results. This paper aims at reducing the relative error between the experimental and numerical results using a lean six-sigma based approach in order to proceed with the next steps of the research project.