This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Gas pumping units (GPU) GTK - 25i produced at Nuovo-Pignon company (Italy) have been in operation for more than 40 years and most of them have fulfilled the established service life or are close to it, and therefore their further operation can lead to emergency situations. Considering that on the Urengoy- Pomary-Uzhgorod gas pipeline today more than 150 GTK - 25i are in operation and the urgent task is to ensure their reliable and efficient operation, which requires the development of diagnostic software - methods and means of diagnosing GTK - 25i and the analysis of the current state of diagnosis of GPA relative to the GTK - 25i showed the lack of effective methods for their diagnosis and indicated the need to develop new methods and means of diagnosing the GTK - 25i on the basis of modern information technologies. The results of the study of 16 statistical characteristics of the technological parameters of the GTK - 25i are given, and 8 of them were selected for further step-by-step discriminant analysis, implemented in the STATISTICA software package, on the basis of which a procedure for selecting a diagnostic sign was developed, which allows to clearly recognize the technical condition of the GPU. The results of the development of the method of diagnosing the GTK - 25i on the basis of artificial neural networks are presented, using as an informative parameter for teaching an artificial network of acoustic noise generated by its axial compressor during operation. The construction of the INS is carried out using the MATLAB software package, in particular, the GUI - Graphical User Interface of the Network Pattern Recognition Tool. When developing diagnostic methods, information was used on the technical condition of the GTK - 25i, which is operated at the compressor station of the Bogorodchansky LPUMG UMG "Prikarpattransgaz", in the form of its technological and vibroacoustic parameters obtained for the repair of the GTK - 25i, after repair and after a period of long-term operation from using the developed information-measuring system.
Gas pumping unit, artificial neural networks, discriminant analysis, statistical characteristics, diagnostic method, experimental data, algorithm
The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.
