March 28 ~ 29, 2025, Virtual Conference
Ilgar Aliyev, Head of Operation and reconstruction of buildings and facilities Department, Azerbaijan University of Architecture and Construction, Baku, Azerbaijan
The management systems for the operational processes of existing parallel gas pipelines under non-stationary conditions are inefficient and technically complex. Additionally, the outdated elements, which are physically worn and do not meet the requirements of modern control technologies, fail to react quickly enough and only report problems after a significant amount of gas is lost to the environment. For this reason, it is crucial to develop new technological schemes that adopt modern control methods in the reconstruction phase of existing pipeline systems in order to eliminate negative outcomes. In this work, based on the analysis of the mathematical model of non-stationary gas flow in parallel gas pipelines, an algorithm based on machine learning technology was developed for the information database of the control and management point of technological processes. The developed algorithm can be recommended for solving engineering technology problems related to the prediction of technical maintenance and management of emergency situations in pipelines. Additionally, in order to ensure the continuity of gas supply to consumers during emergency situations, an empirical formula for determining the locations of connecting pipes in real-time mode based on the analysis of data from pressure sensors was derived. The application of this formula at the central control point allows for the implementation of a remote activation system for the connecting pipes. The analysis reveals that the pressure variations at the beginning and end of the parallel pipeline are directly dependent on the activation time of the automatic valves. This serves as the basis for obtaining information at the control point regarding the activation of the valves on the pipeline.
Pressure sensor, separating devices, valve, connecting pipe, technological scheme, algorithm.