Optimization of the structure of turbine blades produced by methods of additive technologies

Cover Page

Cite item


This study is devoted to the analysis of methods for optimizing the process of manufacturing turbine blades using additive technologies, taking into account the factors of external and internal influence on the finished product, determining the parameters for manufacturing the part, as well as the technical and technological aspects of additive manufacturing of the part. The validity of this integrated approach lies in the complexity of production and the extreme nature of the operation of the part itself, which makes it impossible to accurately estimate the duration of the operation period. But the ability to additively manufacture turbine blades with precise final characteristics will allow the production of a high-quality part with a predictable operational process. In view of these requirements, the authors give recommendations on the criterial and algorithmic support of the process of optimizing the manufacture of a gas turbine engine blade. As a result of the research, it was concluded that the main criterion for optimizing the shape of a gas turbine engine blade is to maintain a constant distance between the corresponding boundary points of the blade sections. Therefore, it will be more efficient and expedient to optimize not the shape of the blade, but the composite from which it is made.

Full Text

Introduction Additive technologies are gaining more and more popularity in the production of parts that are subject to high quality control and durability require- ments. The whole process of additive manufacturing should be monitored at each stage of material-production-post-processing-product to determine the most optimal parameters. Turbine blades have a complex geometric shape, at the same time, high requirements are imposed on the quality of their manufacture, since a violation of the geometric shape can lead to a decrease in the efficiency of the turbine. Quality control of blades is a multi-stage process, which in the vast majority of cases is carried out manually, whereas at present methods of optimizing production processes and quality control of machine-building parts by mathe- matical modeling are becoming more and more re- levant [1-4]. 1. Statement of the research task The purpose of the work is to determine the optimal characteristics of the additive manufacturing process of a gas turbine engine (GTE). At the first stage of developing the technology of additive manufacturing of blades, it is necessary to develop criteria that must be met during the production of the part to achieve the maximum degree of quality. First, it is necessary to identify several control parameters of reliability. Since reliability is a time-varying parameter, the group of control parameters should be exclusively dynamic in nature and have the ability to transform, which can be tracked and recorded, while highlighting the influence of a specific factor on the product under study [5]. The first and most obvious control parameter is temperature. It should be noted that temperature parameters, in particular fluctuation fields, are quite difficult to measure and track at high temperatures. Therefore, one of the objectives of the study is to assess changes in the nature of the propagation of the temperature field during the life cycle of the part. The next equally important parameter is the response to vibration - the vibration spectrum. Measu- ring this parameter is also a technically difficult operation, since this spectrum must be taken on a moving apparatus, the mounting of sensors of which does not meet the necessary reliability. Technically, the system includes a sensor equipped with a microwave element. The response to the vibration effect is formed inside the sensor itself, namely, by converting mechanical vibrations into high-frequency radio signals. This process consists of the following steps [6]: - excitation of the sensor receiver when the blade moves; - detection of mechanical vibration waves; - translation of mechanical vibration into radio frequency vibrations. As in the case of assessing the detail by temperature fluctuations, in this situation it is also ne- cessary to identify diagnostic signs. In this case, it is recommended to consider by analogy not one parameter, but a pair: a spectrum describing the state of the blade under the influence of vibration and anomalies in the structure of this spectrum, which will indicate critical states of the blade that can lead to breakage. Any software product can be used for visualization and calculation, which will make the following assumptions [7]: - replacement of metal elements with equivalent electric currents; - excitation of the medium in the chamber vo- lume by these currents; - imposition of boundary conditions; - splitting of conducting surfaces into elementary platforms, the length of which does not exceed L = λ/8, where λ is the wavelength of the electromagnetic field; - compilation of systems of linear algebraic equations for these currents; - solution of the obtained equations. Such a significant number of assumptions, at first glance, complicates the calculation and modeling, but these assumptions allow us to move away from the practical development of the system, having a mini- mal set of initial data; simulate the remaining parameters, or immediately obtain oscillatory characteristics by setting only qualitative parameters of the blade. The proposed method makes it possible to simulate the oscillation spectrum fairly accurately and without the use of additional means [8]. In contrast to the simulation of vibration spectra, diagnostics using anomalies in the structure of the spectrum is a more analytical issue, requiring an understanding of the life cycle of the blade [9]. For the analytical evaluation of the data obtained, it is necessary to identify criteria in accordance with which anomalies should be compared and calculated. Due to the fact that we remove 2 wave characteristics - working (vibration) and radio frequency (converted by the sensor), then the criteria will be set for these measured values. It is also necessary to take into account that the operating characteristic has such a property as the cyclicity of responses, which must be taken into account when calculating and modeling more than one cycle. Then, according to [6], the frequency of the workflow is (1) where Fr - the number of revolutions (rpm) of the rotor; FB - the rotation frequency of the rotor, Hz; n - the number of turbine rotor blades. Let us consider the use of the spectrum of radio response sequences. The workflow with this spectrum modulates the radio frequency oscillation in the autodyne sensor (Figure 1). When studying the figure, it becomes obvious that the processes of vibration of the gas turbine engine during movement and the flow of the medium passing through the gas turbine engine contribute to each peak section. The graph shows that two characteristics of the read signals are displayed at once: the ampli- tude for the vibration effect and the frequency for the translated radio signal. Figure 1. Radio frequency response spectrum, hHz Since the spectral components in this case are of interest not only as an amplitude characteristic, but also as a true value modulo, it is necessary to develop a mathematically sound calculation method, according to which the main control parameters of reliability for any given blade will be calculated further. All calculations and algorithms will be further carried out within the (-θ1, 0) half-wave segment of the cosine, where , t are dimensionless [9]. The approximation of the symmetric branch of the cosine should be carried out taking into account the slow: rt(r < 1) and fast oscillation phases: qt < (q < 1) having an amplitude of µ << 1. Based on the above, we will make a system of equations: (2) Since there is a radio frequency characteristic in the spectral characteristic under consideration, the use of Fourier coefficients will be legitimate: (3) A significant advantage of the chosen approximation method is the simplicity and possibility of obtaining equations without composing systems and matrices, which not only simplifies calculations, but also reduces the probability of error. The main components of the study and approximation are the constant component (4) and the amplitude of the first harmonic (5): . (4) (5) In the process of this study, the determination of the parameters of the working spectrum will allow not only to evaluate the characteristic parameters of the geometric and physico-chemical properties of the working blade, but also by the presence of a discrete component of the spectrum to determine additional effects from the entire unit of equipment on a single blade (Figure 2). This factor is often not taken into account in the workflow, while providing a huge impact on the durability of the blades [10]. Figure 2. The working spectrum of the GTE blades, kHz The main diagnostic parameter is the vibration spectrum. Let us consider in more detail its graphi- cal arrangement (Figure 3), in which the initial indications of the movements of control points on the surface of the blade are translated into a spectrum by differentiation. Usually, this spectrum has a non-smooth envelope even in a safe situation, and nevertheless, it is it that is taken as a standard when compared with any real spectrum when evaluating performance characteristics. Figure 3. Vibration response spectra, Hz: 1 - reference; 2 - abnormal; 3 - critical; 4 - emergency Table 1 shows the parameters used to control the reliability of blades, modeling, and calculations in the manufacture of blades, including using additive technologies. Diagnostic responses and their causes No. Cause Response 1 Shaft speed Change the workflow interval 2 Shaft beat Component FB on several of the impellers 3 Deformation of the blades Change in the value of 4 Vibration of the blades Blurring of spectral lines 5 Blade breakage Failure of the blades of the FB component, but of smaller am- plitude and observed on one of the impellers The table shows the most significant criteria that can be used for work on the development of additive manufacturing technology parts. In practice, the assessment of the node's condition is based on a comparison of the readings of several sensors in the corresponding measuring channels. However, excessive complication by introducing additional parameters can lead to an increase in operating time, as well as to errors [7]. 2. Choosing an optimization method When choosing an optimization method, first of all, it is necessary to be guided by two important factors. Firstly, work is carried out on a product of a rather complex geometric shape, characterized by a significant set of requirements for physical and mechanical properties. Secondly, the optimization of the part must be carried out in the context of its manufacture using additive technology. Thus, re- strictions will be set not only taking into account the operational properties of the part, but also the method of its manufacture [11]. Since the task is multifaceted, it is possible to use the Pareto principle or transform a multi-criteria design problem into a single-criteria problem. At the first stage, a vector model is created that will allow you to evaluate the shape of the part and make possible adjustments [12] (Figure 4). Анализ лопастей турбины ГТД из карбида титана в пакете Ansys | Статья в журнале «Молодой ученый» Figure 4. Creation of a vector model of a GTE blade After building the model, focusing on the final goal, it is necessary to determine the optimization method, which depends on the method of setting parameters, the dimension of the parametric space, the smoothness of the goal functions and constraints, a constant and variable set of restrictions on iterations, etc. These operations are performed using various software modules, since it is quite difficult to manually account for all parameters, and, in addition, the software package allows you to save all iterations and results [13]. The choice of the software product is also deter- mined by whether the optimization is local or global, the choice of the method and the limits of approximation of constructions and transformations will depend on this. Of course, software products that work with local parameters solve problems more quickly, but they are less loaded with source data [14]. Local approximation methods use current information about the point and, possibly, the results of previous iterations. Based on this information, the transition to the next point is carried out. If the problem has several local optima, then each of them will have its own area of attraction, which may be characterized by a complex structure. In this case, global approximation methods have some advantage, especially if the found approximation corresponds to the true functions of the goal and constraints. Despite the fact that the construction of a global approximation requires significantly more calculations of goal functions and constraints, finding several solutions or a Pareto set for a multi-criteria task will be carried out without additional calls to an external calculation program (Figure 5). Methodically, the optimization of the blade design cannot be reduced to improving a specific parameter without studying the impact This parameter is used in a real system both in a static and dynamic state [15]. Видеоурок CADFEM VL1416 - Работоспособность лопатки при наличии трещины с учетом пластичности - YouTube Figure 5. Selection of areas for local approximations 3. Discussion In general, it is possible to describe and justify an algorithmic approach for the software optimization of a structural element - the blades of a gas turbine engine [16]. Like any modeling software package, this product must have a block or modular structure to divide the task into post-evaluated processes, the results of which in the final module are reduced to obtaining the desired characteristic. A special control program can be used to control the interaction between modules (Figure 6). For simplification, such software packages often use dialog boxes together with a graphical interface that reflects not only the final, but also intermediate results. Figure 6. Structure of interaction between modules in the optimization problem The calculation process itself can take place using the finite element method or the boundary element method [17]. When executing this command file, the following is produced [18]: - building a geometric model; - building a finite element grid; - application loads; - calculation of stress-strain states; - output of results. The algorithm of the program is always the same, so even a single change in parameters triggers the same data processing mechanism. The study of practical data and equipment catalogs allowed us to conclude that the main approaches to optimizing the blades of the gas turbine engine are reduced to solving the problems of reducing the mass of the entire structure (Figure 7) and obtaining a stress-strain state in the blade and disk that meets the requirements and limitations on manufacturabi- lity and strength, which refers to the task of designing the shape of the part [19]. When developing a method for optimizing the blade design, it is necessary to solve not only the problem of geometric characteristics, but also to preserve or increase the reliability and wear resistance of the blade [20]. One of the key points of modeling is the evaluation of the aerodynamics of the part being developed. To conduct research, a part in a computer model is placed in a stream and the emerging pattern of vortex flows and flow around the blade is evaluated (Figure 8). лопатка_АКТО.jpg Figure 7. Reduction the mass of the structure due to the creation of hollow blades Опыт использования ANSYS CFX при доводке конструкции лопаток турбины авиационного двигателя Figure 8. Aerodynamic evaluation of the model The ultimate goal is to obtain a blade shape that will not overload the system with its mass but will remain strong enough and resistant to vibration loads, temperature fluctuations and will maintain the trajectory of movement in the flow. To fulfill these requirements, it is necessary to take as an optimization criterion the preservation of the constant distance between the corresponding boundary points of the blade sections, especially regarding to the sections on the input and output edges of the blade [21]. It means, it is necessary to optimize the shape of the blade itself so that it remains for a long period of operation [22]. Therefore, it will be more efficient and expedient to optimize not the shape of the blade, but the composite from which it will be made. Additive technologies provide such an opportunity. Taking into account the ability of additive technologies to work with almost any metals and alloys, the proposed optimization direction is extremely relevant [23]. Conclusion Analysis of the effect of control parameters on the reliability characteristics of the blade of a gas turbine engine showed that as an optimization criterion, it is necessary to take a constant distance between the corresponding boundary points of the blade sections to preserve the shape of the blade over a long period of operation. In the process of the work, a conclusion was made about the effectiveness and expediency of optimizing not the shape of the blade, but the composition of the composite from which it will be made. A method for optimizing the shape of a GTD blade manufactured using additive technology is proposed.

About the authors

Vladislav O. Eroshenko

Peoples’ Friendship University of Russia (RUDN University)

Author for correspondence.
Email: vladrusty00@yandex.ru
ORCID iD: 0000-0002-3334-7241

postgraduate, Department of Mechanical Engineering Technologies, Academy of Engineering

6 Miklukho-Maklaya St, Moscow, 117198, Russian Federation

Marianna Yu. Malkova

Peoples’ Friendship University of Russia (RUDN University)

Email: malkova-myu@rudn.ru
ORCID iD: 0000-0002-6939-1658

Doctor of Technical Sciences, Professor, Department of Mechanical Engineering Technologies, Academy of Engineering

6 Miklukho-Maklaya St, Moscow, 117198, Russian Federation

Alexander N. Zadiranov

State Fire Academy of EMERCOM of Russia

Email: zadiranov@mail.ru
SPIN-code: 2873-6465
Doctor of Technical Sciences, Professor, Department of Combustion Processes and Environmental Safety, Educational and Scientific Complex of Combustion Processes and Environmental Safety 4 Borisa Galushkina St, Moscow, 129366, Russian Federation

Alexey V. Meshcheryakov

State Fire Academy of EMERCOM of Russia

Email: malviktpp@gmail.com
ORCID iD: 0000-0001-6620-8590

Candidate of Technical Sciences, Associate Professor, Department of Combustion Processes and Environmental Safety, Educational and Scientific Complex of Combustion Processes and Environmental Safety

4 Borisa Galushkina St, Moscow, 129366, Russian Federation

Siamak Ghorbani

Peoples’ Friendship University of Russia (RUDN University)

Email: gorbani-s@rudn.ru
ORCID iD: 0000-0003-0251-3144

Candidate of Technical Sciences, Associate Professor, Department of Mechanical Engineering Technologies, Academy of Engineering

6 Miklukho-Maklaya St, Moscow, 117198, Russian Federation


  1. Wang Yu, Denisov OV, Denisova LV. Modeling of the thermal control system of a nanosatellite using loop heat pipes in orbital flight. RUDN Journal of Engineering Research. 2021;22(1):23-35. (In Russ.) http://doi.org/10.22363/2312-8143-2021-22-1-23-35
  2. Antipov YA, Shatalova II, Shkarin KV, Lapin MV, Sokolov DA, Grinin AO, Toptygin KP. Features of modeling a highly efficient multi-stage steam compression heat pump unit. RUDN Journal of Engineering Research. 2021;22(4):339-347. (In Russ.) http://doi.org/10.22363/2312-8143-2021-22-4-339-347
  3. Mamaev VK, Vinogradov LV, Oshchepkov PP. Modeling of the lattice of profiles of a transport gas turbine engine. RUDN Journal of Engineering Research. 2019;20(2):140-146. (In Russ.) http://doi.org/10.22363/2312-8143-2019-20-2-140-146
  4. Kornilova AV, Zayar Ch. Determination of permissible parameters of defects in basic parts of forging and pressing machines. RUDN Journal of Engineering Research. 2019;20(4):308-315. (In Russ.) http://doi.org/10.22363/2312-8143-2019-20-4-308-315
  5. Galiev VE, Fatkullina DZ, Tamindarov DR. On the problems and prospects for the manufacture of compressor blades. Herald of the Bauman Moscow State Technical University. 2014;(FS77):10-25. (In Russ.)
  6. Tarasenko IuP, Levanov IuK, Tsareva IN, Shchegolev I L, Chernikov A I. Assessment of the operational condition of the HDPE blades of the GTK-25I unit and technology for extending their life. Gas Turbine Technologies. 2004;(7):26-29. (In Russ.)
  7. Golubentsev AV. Increasing the fatigue characteristics of the working blades of the GTU on the basis of improving the technology of casting and heat treatment (dissertation of the Candidate of Technical Sciences). Rybinsk; 2016. (In Russ.)
  8. Andrienko AG, Olshanetsky VE, Sklyarevskaya VN, Shmyrko VI. Estimation of damageability and prediction of durability of gas turbine rotor blades. Technological Systems. 2001;(3):71-74. (In Russ.)
  9. Smyslov AM, Bybin AA, Dementiev AV, Nevyantseva RR, Novikov AV. Evaluation of the maintainability of the blades of gas-pumping units. Thermal Power Engineering. 2011;(2):30-35. (In Russ.)
  10. Orlov MR. Technological support of the resource of the working blades of the first turbine stages of aviation and ground-based gas turbine installations (abstract of the dissertation of the Candidate of Technical Sciences). Moscow; 2008. (In Russ.)
  11. Bhandari S, Lopez-Anido R. Finite element modeling of 3D-printed part with cellular internal structure using homogenized properties. Progress in Additive Manufacturing. 2019;4:143-154. https://doi.org/10.1007/s40964-018-0070-2
  12. Farbman D, McCoy C, Materials testing of 3D printed ABS and PLA samples to guide mechanical design. ASME 2016 11th International Manufacturing Science and Engineering Conference. London: Eurospan; 2016. https://doi.org/10.1115/MSEC2016-8668
  13. Ferro CG, Mazza A, Belmonte D, Seclì C, Maggiore P. A comparison between 3D printing and milling process for a spar cap fitting (wing-fuselage) of UAV aircraft. Procedia CIRP. 2017;62:487-493.
  14. Kazemian A, Yuan X, Cochran E, Khoshnevis B. Cementitious materials for construction-scale 3D printing: laboratory testing of fresh printing mixture. Construction and Building Materials. 2017;145:639-647.
  15. Liljenhjerte J, Upadhyaya P, Kumar S. Hyperelastic strain measurements and constitutive parameters identification of 3D printed soft polymers by image processing. Additive Manufacturing. 2016;11:40-48.
  16. Mahshid R, Hansen HN, Højbjerre KL. Strength analysis and modeling of cellular lattice structures manufactured using selective laser melting for tooling ap- plications. Materials and Design. 2016;104:276-283. https://doi.org/10.1016/j.matdes.2016.05.020
  17. Raj SA, Muthukumaran E, Jayakrishna K. A case study of 3D printed PLA and its mechanical properties. Materials Today: Proceedings. 2018;5:11219-11226.
  18. Yuen PK. Embedding objects during 3D printing to add new functionalities. Biomicrofluidics. 2016;10(4):1-10.
  19. Song Y, Li Y, Song W, Yee K, Lee KY, Tagarielli VL. Measurements of the mechanical response of unidirectional 3D-printed PLA. Materials and Design. 2017;123: 154-164. https://doi.org/10.1016/j.matdes.2017.03.051
  20. Yao T, Ye J, Deng Z, Zhang K, Ma Y, Ouyang H. Tensile failure strength and separation angle of FDM 3D printing PLA material: experimental and theoretical analyses. Composites Part B Engineering. 2020;188:107894. https://doi.org/10.1016/j.compositesb.2020.107894
  21. Wang L, Zhang M, Bhandari B, Yang C. Investigation on fish surimi gel as promising food material for 3D printing. Journal of Food Engineering. 2018;220:101-108.
  22. Lee W, Jeon S, Kim D. 3D-printed micromixer with helical blades for highviscosity fluids. Department of Chemical Engineering. 2015. Available from: http://www.rsc.org/images/LOC/2015/PDFs/Papers/1253_T.320e (accessed: 12.08.2022).
  23. Pastor-Artigues M-M, Roure-Fernández F, Ayneto-Gubert X, Bonada-Bo J, Pérez-Guindal E, Buj-Corral I. Elastic asymmetry of PLA material in FDM-printed parts: considerations concerning experimental characterisation for use in numerical simulations. Materials. 2020;13(1):15.

Copyright (c) 2022 Eroshenko V.O., Malkova M.Y., Zadiranov A.N., Meshcheryakov A.V., Ghorbani S.

License URL: https://creativecommons.org/licenses/by-nc/4.0/legalcode

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies