Machine-Learning-Based Condition Assessment of Gas Turbines—A Review

Condition monitoring, diagnostics, and prognostics are key factors in today’s competitive industrial sector. Equipment digitalisation has increased the amount of available data throughout the industrial process, and the development of new and more advanced techniques has significantly improved the performance of industrial machines. This publication focuses on surveying the last decade of evolution of condition monitoring, diagnostic, and prognostic techniques using machine-learning (ML)-based models for the improvement of the operational performance of gas turbines. A comprehensive review of the literature led to a performance assessment of ML models and their applications to gas turbines, as well as a discussion of the major challenges and opportunities for the research on these kind of engines. This paper further concludes that the combination of the available information captured through the collectors and the ML techniques shows promising results in increasing the accuracy, robustness, precision, and generalisation of industrial gas turbine equipment.

Multi-Criteria Modeling of Processes in a Vaporjet Condensation Turbine Ejector Mixing Chamber and Its Design

The paper presents recommendations on designing an efficient mixing chamber and regulating air and steam flows in the elements of a steam-jet apparatus based on the results of multi-criteria modeling of gas-dynamic pro-cesses. The paper also reveals the possibilities to provide a greater reduction in losses, air suction into the turbine and equipment repair costs compared to the cases when individual criteria for increasing the steam turbine equipment efficiency are used. The selected calculation methods, i.e. innovative (multi-criteria) and standard (standard), have been analyzed comparatively. As a result of the methods evaluation, the authors se-lected the following indicators: increased energy efficiency of the turbine due to the prevention of air suction in the turbine seals; increased efficiency of condensing turbines; reduced repair and maintenance costs for the turbine and ejector due to the ejector mixing chamber cross-sectional area optimization. Based on the comparison, the best results were shown with the multi-criteria modeling according to the indicated parameters, the coefficients of the flow rate, working steam consumption, size reduction and increase in service life from 20 to 40 years. The innovative method requires a larger number of optimality criteria and the regulation of process pa-rameters and design geometry based on direct and feedback multi-criteria models. Using multi-criteria modeling, the authors determined the necessary composition of optimization parameters, i.e. the flow area of the designed mixing chamber, steam flow rate and thermodynamic flow parameters, as well as suction pressure.

Assessment of ability to varnish turbine oils

The reasons for the formation of varnish deposits of turbine oils are determined. It is shown that the performance of turbine oils is determined by methods of ASTM D7843 (MPC) and ASTM D6971 (RUL) can be assessed, the process of formation of varnish deposits, the level of antioxidant properties, antioxidant content of, amend, and extend the oil change, and to predict critical situation in the operation of turbine equipment. The mechanism of formation of varnish deposits in turbine oils is proposed.

APPLICATION OF PVD COATINGS TO INCREASE THE SERVICE LIFE OF GAS TURBINE MINI HEAT POWER PLANTS WITH IMPROVED ACOUSTIC PERFORMANCE

this article deals with the problem of the influence of gas turbine equipment on its energy efficiency, environmental performance and acoustic properties of thermal mini heat power plants. Environmental aspects of the introduction of gas turbine technologies in heat and power industry are very important. The paper considers technological limitations of the wide use of gas turbine equipment and ways to solve them. The article discusses the developed original technology to improve the performance characteristics of parts that have exhausted their service life of aviation gas turbine engines, with the aim of using them as heat installations in heat and power engineering with improved acoustic performance. Constructive solutions are proposed for upgrading the serial installation of PVD coatings, for the possibility of applying nanostructured films according to the technology proposed in the article, with an increase in its economic and environmental performance.

Examination of informativeness of diagnoses expressed with multiple-valued logic

This paper presents the essence of an examination of informativeness in the diagnostic information outputs expressed with multiple-valued logic. The diagnostic test required for the examination was completed on wind turbine equipment. The examination included a constant set of determined diagnostic output values. The DIAG 2 diagnostic system was used for the examination and the diagnostic test. DIAG 2 is a smart diagnostic system capable of any inference k of the set {k = 2, 3, 4}. The examination results were expressed in an Object State Table, separately for each k-valued logic of inference tested. Keywords: technical diagnostics, diagnostic inference, multiple-valued logic, artificial intelligence

Diagnosing of a complex technical object in four-valued logic

This paper presents the essence of an investigation of a complex technical object with the use of four-valued logic. To this end, an intelligent diagnostic system (DIAG 2) is described. A special feature of this system was its capability of inferring k at {k = 4, 3, 2}, in which case the logic {k = 4} is applied. An important part of this work was to present the theoretical foundations describing the essence of inference in the four-valued logic contemplated. It was also pointed out that the basis for classification of states in the multiple-valued logic of the diagnostic system (DIAG 2) was the permissible interval of changes in the values of diagnostic signal features. Four-valued logic testing was applied to a system of wind turbine equipment. Keywords: technical diagnostics, diagnostic inference, multiple-valued logic, artificial intelligence