IMPROVING SELF-DIAGNOSTIC METHODS OF FLOW MEASUREMENT SYSTEMS BASED ON ARTIFICIAL INTELLIGENCE

The article examines methods for improving self-diagnostics of consumption measurement systems based on artificial intelligence in the context of industry digitalization and the development of cyber-physical systems. It has been shown that traditional flow meters used to measure the flow rate of liquids and gases are subject to mechanical, hydraulic, electronic, and hidden failures, which reduce the accuracy and reliability of measurements. A justification for the need to transition from classical maintenance methods to intelligent self-control methods that ensure the detection of anomalies and hidden malfunctions in real time is presented. A multi-level architecture of intelligent self-diagnosis is proposed, including feature extraction modules, soft-sensor models, spectral analysis, and machine learning algorithms. A comparative analysis of various diagnostic methods was conducted, and it was shown that the integration of AI models allows for an increase in failure detection accuracy to 90-95%. Conclusions have been drawn regarding the prospects of applying intelligent flow measurement systems within the framework of Industry 4.0 and 5.0 concepts, as well as in the tasks of predictive analysis and ensuring the functional safety of industrial processes.

Hali tarjima qilinmagan