An algorithm has been developed that will significantly speed up the diagnostics of gas pipelines in the Unified Gas Supply System.

Scientists from the Gubkin Russian State University of Oil and Gas (National Research University) have developed an algorithm that reduces the processing time of data obtained during in-line inspection of gas pipelines by dozens of times – from several hours to minutes.

Pilot operation of the system at facilities in the Unified Gas Supply System of Russia (UGS) is planned for the end of 2026.

The method developed by the scientists allows for the automated decoding of information obtained from in-line flaw detectors – instruments used to inspect the condition of gas pipelines. It is based on a mathematical model and a software algorithm that automatically identifies welded joints on magnetograms, analyzing the data even in the presence of interference. As a result, the operator receives a final result, eliminating errors that can occur during manual work and reducing labor costs.

"Traditionally, processing one section of a main gas pipeline requires several hours of painstaking work by an expensive specialist. Up to 10% of the data was entered with errors. Now, this process will be reduced tenfold by using an automated processing algorithm. We have not only accelerated the procedure but also reduced the impact of human error by transferring routine tasks to a computer program. This is an important step toward creating a risk forecasting system based on technical condition analysis for the entire pipeline infrastructure of the country," said Konstantin Zhuchkov, Associate Professor of the Department of Thermodynamics and Thermal Engines at Gubkin University.

In the next stage of the project, the development is planned to be supplemented with neural network technologies.

"Essentially, a diagnostic specialist will very soon become the operator of a fully robotic system that will transmit diagnostic reports in near-real time directly from the pipeline," the scientist added.

Accelerating the processing of diagnostic data will optimize the maintenance costs of a gas pipeline system over 200,000 kilometers long, according to the project participants.

"Cost savings in diagnostics and repairs will help curb rising gas prices for households and industrial enterprises," says Konstantin Zhuchkov.

In the future, the scientists plan to develop machine learning methods to predict the risk of damage to UGSS facilities and develop recommendations for maintaining equipment operability.

The project is being implemented at the UGSS Systems Research Center, established at Gubkin University in 2020 to develop import-independent software for the needs of the domestic gas industry and pipeline gas transportation.