Blade Profile Research Articles

Research on uncertainties in fuel centrifugal pump based on prediction and reconstruction of internal flow field

Geometric machining errors in the blade profile and variable operating conditions in the extreme operating environment are primary factors leading to the uncertainties in pump performance. This paper presents an analysis of uncertainties of fuel centrifugal pumps by modeling the geometry uncertainty in blade machining based on the Karhunen–Loève (KL) expansion and using a polynomial chaos expansion (PCE) model. First, the geometric uncertainty in the blade machining is described by the KL expansion in three sections and a stochastic simulation of the blade geometry is performed. Then, a PCE surrogate model is trained based on the least angle regression method and validated by the bootstrap method to quantify the uncertainties of performance indices. Finally, the influence mechanism and relative importance of each input uncertainty parameter are investigated using a quasi-Monte Carlo simulation method. The results show that the KL expansion of the blade profile uses the random vector perturbation superposition of three stream surface, achieving the dimensional reduction in the blade machining error. The PCE surrogate model, trained with a dataset of 3 × 106 sample points, exhibits excellent fit, and the R-squared and adjusted R-squared for head coefficient and efficiency are both above 80%. The variance of parameter control points of the reconstructed flow field is less than 0.002. The uncertainties in both operating conditions and parameters have an influence on the distribution of the global flow field, while the influence of the uncertainty in machining error on the global flow field mainly concentrates on the power-generating positions of the blade.

The Influence of Blade Wrap Angle on the Internal Flow Field and Hydraulic Performance of Double-Suction Pump for Yellow River Pumping

Abstract Analyzing the influence of blade wrap angle on the internal flow field and hydraulic performance of the pump, five different blade wrap angles were designed. Based on RNG k-ε and DPM were used to conduct a steady-state numerical simulation of impellers under the same conditions and different sand volume fractions, respectively. The results show that as the blade wrap angle increases, the flow separation and vortices in the impeller flow passage also decrease, and the flow is closer to the blade profile, but the friction loss in the impeller flow passage also increases. There is an optimal blade wrap angle to maximize the pump head and efficiency. By increasing the wrap angle of blade, the low-pressure zone at the blade inlet area increases, the high-pressure zone at the outlet decreases, and the low-speed zone within the impeller decreases, increasing the outlet velocity. When the sand volume fraction is less than 5%, the effect of sand volume fraction on the hydraulic performance of the pump is not obvious. When the sand volume fraction is more than 5%, the change of wrap angle of blade has a significant effect on the hydraulic performance of the pump.

Optimization of Savonius rotor blade performance using Taguchi method: Experimental and 3D-CFD approach

The Savonius hydrokinetic turbine (HKT) is both cost-effective and reliable, providing clean energy with minimum environmental impact. The efficiency of the Savonius rotor can be improved by modifying the blade profile to increase the effective torque. However, past research works reported that modifying the blade profile is quite challenging due to small Cp improvement and design constraint. Therefore, the present study proposes a newly developed blade profile blueprint, akin to parameterizable designs such as the modified Bach and Benesh profiles but offering more shape versatility. This new blueprint is optimized with a systematic design of experiment (DOE) using a 3D computational fluid dynamics (CFD) simulation to maximize the Cp. The best design is determined statistically using the Taguchi method and analysis of variance (ANOVA). The optimized rotor profile enhances Cp by approximately 10.9 % compared to the conventional Savonius rotor at the optimal TSR (best Cp = 0.159) and 16.7 % improvement at a higher TSR value of 0.9 (Cp = 0.158). Moreover, the optimized rotor outperforms the Fibonacci rotor (Fibonacci blade profile) by 27 % in terms of efficiency. A thorough CFD analysis suggests that the optimized blade profile has a greater lift generated at the convex side of the advancing blade due to its streamlined shape, which delays the flow separation further up to the root of the blade. The confirmation test (experiment) is also performed to verify the Cp improvement of the optimized rotor, and the results herein concluded that the present CFD prediction is accurate and consistent with the experimental values.

GA-BP neural network-based nonlinear regression model for machining errors of compressor blades

Given that blade machining errors can substantially degrade aeroengine performance and reliability, there is a critical imperative to implement stringent control over blade machining errors during the manufacturing process. Nevertheless, the prevailing complexity surrounding the numerous machining error types and their obfuscated interrelationships impedes the comprehensive control of blade profile errors. Consequently, unveiling the nonlinear relationships between the distinct machining errors represents a task of paramount importance. To uncover these nonlinear relationships, this work pioneers the application of factor analysis and Genetic Algorithm-optimized BP neural networks to enable nonlinear regression for diverse blade machining errors. Specifically, this paper first constructs an orthogonal factor analysis model to reduce the dimensionality of complex, high-dimensional blade machining error data and extract underlying correlations. This analysis reveals three distinct error groups, each dominated by a latent common factor. Building on these factor analysis outputs, GA-BP neural networks are then leveraged to perform nonlinear regression for intra-group errors. Guided by the Maximal Information Coefficients (MIC) between different errors, we identify control error types within each group to serve as network inputs, with other errors designated as outputs. To prevent overfitting and accelerate convergence of the neural network, Monte Carlo method is used to augment the limited raw error dataset. The augmented data trains the GA-BP network, establishing multivariate nonlinear models with control errors as independent variables and other errors as dependent variables. The model accuracy is validated by a separate validation set. The comparisons between the predicted and true values in the validation set reveal that the relative errors are around 5 %, indicating that the models achieve an accuracy of approximately 95 %. These results demonstrate the satisfactory performance of the established models. Overall, this work elucidates previously unclear relationships between distinct blade machining errors. These new insights establish the foundation for comprehensive error control and will directly benefit blade design, manufacturing, and performance.

Improving turbulent corrosion in Francis turbine from aspects of fluid mechanics

Turbulent corrosion within Francis turbines is a pressing concern in hydropower generation. This study investigates hydrodynamic factors contributing to this corrosion phenomenon and proposes strategies for mitigation, enhancing turbine performance and longevity. Employing Computational Fluid Dynamics (CFD) simulations, the study comprehensively analyzes the intricate water flow dynamics within the turbine, identifying areas susceptible to turbulence and guiding design adjustments. Turbulence reduction, achieved through optimizing blade and runner geometries, and the reduction of eddy currents through well-designed components, play pivotal roles in mitigating corrosion. Flow field analysis, turbulence reduction, and eddy current minimization emerge as central components in addressing turbulent corrosion. CFD simulations offer invaluable insights into flow dynamics, while optimized geometries reduce turbulence and associated corrosion risk. Acknowledging limitations, this research primarily addresses hydrodynamic aspects of turbulent corrosion, overlooking complex operational factors. Future research should encompass a holistic perspective by integrating additional variables for a more comprehensive understanding. In design and engineering, while optimizing blade profiles and flow path geometries is essential, exploring advanced materials, protective coatings, and innovative maintenance techniques is warranted. Interdisciplinary solutions should be sought to address multifaceted challenges presented by turbulent corrosion. Adopting a holistic approach and addressing these limitations promises to advance hydropower generation, ensuring safer and more efficient turbine operation.

Modeling of particle erosion of high-pressure turbine based on dynamic mesh method

In this work, we investigate erosion characteristics of a high-pressure turbine by using Oka erosion model coupled with the dynamic mesh method, which allows for the dynamic change of blade profile. The variation of blade profile, material removal and performance degradation during erosion are quantified for the original blade, and the results are compared with those of an optimized blade previously obtained through multi-objective optimization of aerodynamic and erosion resistance performances. It is found that compared with the original blade, the optimized blade is of higher efficiency and lower erosion, suggesting that the previous multi-objective optimization, albeit ignoring the dynamic change of blade profile, still works for the dynamic erosion process. For either blade, the efficiency decreases slowly at first and drops sharply after a period of erosion, because as the erosion proceeds, the regions occupied by the vortex rapidly expand, leading to the blockage of blade passage. The transient erosion characteristics and particle trajectories indicate that the erosion is initially located near the middle and rear of blade on pressure side and the leading edge on suction side; as the erosion progresses, the erosion mainly occurs near the leading edge on pressure side but almost disappears on suction side.

Experimental Studies of Pressure Pulsations in Draft Tube Diffuser of Pump-Turbine Models for Heads up to 200 m

Introduction. Increasing the share of balancing capacities to cover daily peaks in electricity consumption is one of the top priorities of the postwar development of Ukraine’s energy sector.Problem Statement. Today, power plant hydraulic turbines need both to increase effi ciency and to expand their operating range. For example, the new hydraulic units of the Dniester PSP shall operate in turbine mode in the range of 40—100% of rated capacity, while the four previous units operate in the range of 70—100%. This requirement can be met by increasing efficiency and reducing pulsations at low power output.Purpose. Based on studying the infl uence of blade spatial shape of a Francis pump-turbine runners on fl ow parthydrodynamics, to identify the patterns of pressure fluctuations distribution in draft tube diff user of the hydraulic unit model.Materials and Methods. Three options of models (the original and two modifi ed ones) have been studied onthe IMEP ECS-30 hydrodynamic test stand. The runner blades are made of PLA plastic by 3D printing. Pressurepulsations are measured by sensors at two points of draft tube diff user at a distance of 0.2 and 1.5 runner diameters from the runner bottom shroud.Results. Three modifi cations of the pump-turbine runner for heads up to 200 m have been designed and experimentally studied with the use of circumferential lean that diff er from the original version only in relative position of blade profiles. The analysis of obtained energy and pulsation characteristics of the models in turbine modehas shown that the model with the runner having a negative circular blade lean has the best performance. Conclusions. The determined influence of spatial shape of the runner blades on the energy and pulsation characteristics of the Francis pump-turbine model for heads up to 200 m has made it possible to increase its efficiency and to reduce the level of pressure fluctuations in the flow part.

Design and experimental analysis of a new vertical ultra-low-head hydro turbine with the variation of outlet flow level on the head drop section of an open canal

Hydropower generation from ultra-low-head water resources is the salient category. The power can be tapped through different non-conventional turbines and is developing these days with different features. This research aims to develop a technology that applies to ultra-low-head resources from conceptualization to laboratory testing. Among several ultra-low-head water resources, it is concentrated on the power generation from the head drop section of the open canal flow through a new hydro turbine. A vertical shaft turbine with an axial flow type of runner having fixed, flat, and constant blade thickness including fixed guide vanes has been designed conventionally for 0.5 m of head and 0.088 m3/s of discharge. Several experiments have been conducted with fixing the level of flow outlet control from 0.0 m to 0.5 m creating the runner from free to submerged conditions is the novelty of this research. The results obtained for all the independent and dependent variables are presented and discussed for a constant flow of 0.088 m3/s and constant speed of 115 rpm conditions. The uncertainty on measurement was found ±1.85 % and the overall maximum efficiency obtained was 53.8 % for flow 10 % above the design flow. The research has concluded the use of outlet flow control is necessary to achieve the maximum efficiency of the turbine at a specific height and the overall efficiency can be advanced further through the optimization of blade profile.

SIMULATION AND OPTIMIZATION MATERIAL OF WIND TURBINE BLADE PROFILES USING CATIA AND ANSYS,

I Wind energy is a fast-growing way to make power from wind, and it's good for the environment. Wind turbines have big blades that catch the wind to make electricity. This project is all about making those blades work better and cost less money. First, we use special computer programs called CATIA and ANSYS to design and test the blades. CATIA helps us design the shape and size of the blades, while ANSYS helps us see how strong they are and how well they catch the wind. We start by making different blade shapes in CATIA, changing things like length, width, and twist. Then we use ANSYS to test them. We look at things like how much stress they can handle, how much they bend, and how well they catch the wind. After testing, we try to make the blades even better. We use special tools in ANSYS and CATIA to find the best combination of blade features. We want blades that are super-efficient at catching wind, strong enough to last a long time, and don't cost too much to make. By varying material to blade profile checking strength and mode shapes for Aluminium, Carbon fibre and E glass epoxy Keywords: wind energy, Turbines, Wind blades, CATIA, ANSYS

DEVELOPMENT OF PROFILE DEVIATION ANALYSIS METHODOLOGY BLADE PEN OF RESEARCH LOT USING COMPRESSION AND INFORMATION CONVERSION METHODS

As part of the implementation of the concept of the “digital twin” of the Gas Turbine development, a methodology for analyzing deviations in the profile of the fan blade using the example of a research batch using compression and information conversion methods has been proposed. This approach allows detecting incorrectly measured blades, comparing the coordinates of the manufactured profile with those specified in the drawing simultaneously for all controlled sections of the blade, conducting an operational analysis and deciding on the need to take into account the actual geometry of the blades in the calculation models “as manufactured” transmitting the results to the digital platform module for creating digital twins for virtual tests of the actual configuration of the assembly or GTE as a whole.

Aerodynamic performance and starting torque enhancement of small-scale Darrieus type straight-bladed vertical axis wind turbines with J-shaped airfoil

Wind energy is one of the most eminent renewable sources for the generation of power. The increasing enthusiasm toward the advancement of small-scale Darrieus type straight-bladed vertical axis wind turbines (SB-VAWTs) can offer a potential remedy for addressing power shortage and the unpredictability of climate conditions. These particular wind turbines provide distinct advantages over their counterparts due to their linear blade design and uncomplicated structure. However, enhancements are required in their aerodynamic efficiency and self-initiation capabilities. These challenges stem from using traditional straight blade configurations and symmetrical airfoils. By substituting these conventional elements with J-shaped straight blades and along with cambered airfoils, these issues can be effectively overcome. The current study aims to investigate the effect of J-shaped straight blades with a series of cambered airfoils to improve the aerodynamic performance and starting torque of small-scale Darrieus type SB-VAWTs. Therefore, experimental and numerical studies are conducted to analyze the J-shaped airfoil impact with various opening ratios systematically. The J-shaped blade profile is designed by eliminating some portion toward the trailing edge of a conventional airfoil. This analysis demonstrated that the J-shaped blade incorporating a cambered NACA 4418 airfoil outperforms its alternative cambered airfoil designs. The performance of SB-VAWT improves by about 25% by the J-shape of the cambered NACA 4418 airfoil with a 70% opening ratio. Moreover, the use of J-shaped airfoils enhances the self-starting torque of SB-VAWT compared to conventional airfoils.

Enhancing Turbine Blade Manufacturing through MEMS-Based Milling Monitoring

Manufacturing steam turbine blades with intricate shapes poses a substantial challenge. These blades significantly impact turbine efficiency, reliability, and productivity. Their precise formation through milling processes demands vigilant monitoring, especially concerning the cutting tool’s condition, which is responsible for shaping the blade profiles and ensuring high-quality outcomes. Direct tool monitoring, however, disrupts productivity, prompting the need for an indirect Tool Condition Monitoring (TCM) system. Such a system becomes essential for detecting tool wear and damage early, ensuring dimensional accuracy and surface smoothness. This experiment monitors tool conditions by analyzing vibrations generated by an endmill cutter while machining martensitic stainless steel (MSS) AISI 420. The setup employs a cost-effective MEMS vibration sensor, the ADXL 345, and Raspberry Pi for real-time online TCM functionality as the signal processor. This study delves into the sensor’s ability to capture vibrations representing actual tool conditions, focusing on the affordability and effectiveness of MEMS-based monitoring. The successful real-time implementation of MEMS-based monitoring could be an easily accessible gateway for manufacturing industries to embrace cloud-based monitoring systems, aligning with current technological trends. This research aims to underscore the viability and potential adoption of affordable MEMS sensors for comprehensive tool condition monitoring, offering a seamless transition toward cloud-integrated monitoring systems for manufacturing industries.

Multi-fidelity CFD for obtaining the runner blade profile parameters of a Francis turbine for optimum hydrodynamic performance

The high fidelity Navier Stokes equations based CFD codes precisely take care of the turbulent and viscous effects of complex flows passing through the hydraulic turbines. However, for preliminary design synthesis and analysis purposes, this full 3D flow can be assumed as a combination of two 2D flows (meridional axi-symmetric flow and blade-to-blade or cascade flow) which is the basis of low fidelity potential flow theory for hydraulic turbines. This paper presents the utilities of high as well as low fidelity CFD analyses for obtaining a runner blade design which yields an optimum hydrodynamic performance of a prototype Francis Turbine. The complete flow domain of the Francis turbine has been considered for the high-fidelity analysis which is carried out by using ANSYS CFX 16.0 for different operating conditions. Whereas, for the low fidelity investigations, full three-dimensional flow domains of different blade rows (i.e., stay vanes, guide vanes and runner) of the Francis turbine is divided into two 2D flow domains (i.e., an axisymmetric annulus of the turbine space for meridional flow analysis and cascades at different sections from hub to shroud for blade-to-blade analysis). By using the low fidelity analysis, runner blade design parameters have been adequately varied in order to obtain a population of different promising runner blade designs. Performances of these promising designs are evaluated, and the obtained results are thoroughly analysed through the response surface analysis. An optimized hydrodynamic performance design of the runner is obtained for the target values of responses (here, swirl velocity at trailing edge of the runner (Cu2) and stagnation pressure at trailing edge of the runner (P02/ρ)) at different sections from hub to shroud by determining the design parameters of blade profiles at different sections. The adopted multi-fidelity CFD approach is found very effective and has the potential to yield reasonably accurate hydraulic designs of the Francis turbine in considerably lesser time and cost.

Conformally Decambered Natural Laminar Flow Blades for Vertical-Axis Wind Turbines

A two-element natural laminar flow airfoil—commonly used on medium-altitude, long-endurance, unmanned air vehicles—was conformally decambered for application on a two-bladed, H-rotor, vertical-axis wind turbine. Blade kinematics were used to determine the virtual camber line, which was then used to conformally map the original profile onto the chord line. Both decambered and original blade profiles were evaluated experimentally using large chord-radius ratios (0.6 and 0.75) that exploited dynamic stall to produce the driving torque. Decambered blades showed substantially greater power and torque coefficients than the original blades, up to 60 and 27%, respectively, which represents the first experimental validation of conformal decambering. Relatively large peak power coefficients of 0.28 were attained, despite maximum chord-based Reynolds numbers being less than 2×105. Depending upon the chord-radius ratio, either light or deep dynamic stall occurred in the second upstream quadrant, and the flap flow remained attached virtually throughout. In contrast, on the original profiles, massive separation was observed on the blades, and the flap flow remained separated due to assumed outer surface flow separation. Future research should consider surface pressure and flowfield measurements, significantly higher Reynolds numbers, and variable intracycle flap deflection mechanisms to optimize performance and minimize unsteady loads.

Recurrent multi-view collaborative registration network for 3D reconstruction and optical measurement of blade profiles

Aligning multiple viewpoints to reconstruct a complete profile is crucial for the three-dimensional optical measurement of blade profiles. As foundational solutions, rigid pairwise point cloud registration methods often struggle to produce satisfactory results due to significant cumulative errors stemming from pairwise registration errors, while current multi-view registration methods also face challenges when their prerequisites conflict with the practical measurement scenarios of blade profiles. To address these issues, this study proposes a learning framework to recover a transformation vector between the coordinate frames of both the optical sensor and the rotational axis in our developed optical measurement system based on extracted adaptive feature embeddings from the initial coordinate representation of viewpoints, which simultaneously aligns all viewpoints into a complete profile without relying on any prerequisites. Furthermore, an elaborate recurrent updating strategy is incorporated into the framework that enables incremental refinement of the transformation vector, thus aiding the framework in meeting the high-precision measurement requirements at the 0.01 mm level. Experimental results from both theoretical and real-world data, conducted on three representative blades and compared against nine state-of-the-art algorithms, consistently demonstrate the superiority of the proposed method in terms of accuracy, time efficiency, and robustness. The source code is available at https://github.com/Clarkxielf/Recurrent-Multi-View-Collaborative-Registration-Network.

A shock loss reduction method for the low-pressure rotor of a vaneless counter-rotating turbine using circulation redistribution

This paper proposes the circulation redistribution for the low-pressure (LP) rotor of a highly-loaded vaneless counter-rotating turbine (VCRT) to reduce the LP shock losses. The method is to decrease the mid-span circulation and increase the tip circulation at the exit of the LP rotor with unchanged high-pressure (HP) and LP power output. The LP Blade-to-Blade flow field can be considered as a Laval nozzle model. And the high static pressure drop from the throat to the exit results in strong trailing edge (TE) shocks near the LP mid-span. Also, the suction side TE shock (ST) and the reflected shock of the pressure side TE shock (RPT) near the LP mid-span merge into a relatively strong shock very close to the TE. These two factors result in high shock losses. To reduce the shock losses under the design condition, decreasing the stagger angle and the blade exit angle near the LP mid-span are adopted to reduce the LP mid-span exit circulation. This changes the throat area to exit area ratio near the LP mid-span, prompting the actual static pressure drop from the throat to the exit to be reduced and closer to the theoretical value, so that the TE shock strength is reduced. Besides, the intersection of the RPT and the ST is delayed owing to the improved flow angle distributions near the TE, further contributing to the LP mid-span shock loss reduction. Since the LP power output is decreased and the LP tip-leakage flow (TLF) losses are far lower than the HP TLF losses, increasing the stagger angle and the blade exit angle in LP tip regions are employed to increase the LP exit circulation and decrease the HP exit circulation in tip regions. Then the HP and LP efficiency are both raised with unchanged HP and LP power output after optimizing the blade profiles near the mid-span and in tip regions of the LP rotor. The remarkable thing is that the TE shock strength in LP tip and root regions are both reduced after improved because of the decreased ratio of the exit area to the throat area. And the intersection of the RPT and the ST in LP tip and root regions are also both delayed, leading to reduced shock losses. The reduced TE shock losses along the whole LP span appreciably raise the LP and VCRT efficiency by 0.96% and 0.32% respectively.

Measuring the profile of aircraft engine blades using spectral confocal sensors

The geometric parameters of aircraft engine blades need to be precisely measured to ensure the quality of the blades for the normal operation of aircraft engines. This study aims to address the challenges of existing measurement systems in balancing efficiency, accuracy, and completeness. Additionally, it aims to enhance the accuracy and robustness of the algorithm for extracting blade profile characteristic parameters. In this paper, a spectral confocal sensor is employed to establish a blade profile measurement system. The design includes a probe sampling strategy, and a standard ball is used to calibrate the sensor probe’s light emission direction and the precise rotation center of the turntable. The paper proposes the use of methods such as partition search algorithm, binary search, and curvature segmentation to process point cloud data of blade body and tenon. We have conducted experimental measurements on the blades of an aircraft engine. The acquired three-dimensional point cloud data of multiple sets of blade cross-sections and dovetail sections were processed. After calculation, the maximum measurement errors for chord length, maximum blade thickness, tenon width, bottom height, top angle, and bottom angle are −0.0036 mm, −0.00721 mm, −0.0102 mm, −0.00928 mm, 0.0086°, and −0.0058°, respectively. This process validates the effectiveness of the proposed method and has high measurement accuracy. Compared with the CMM method, this method is more accurate in measuring small pits and large curvature micro surfaces, with higher measurement integrity and higher measurement efficiency.

Study on blade zoning measurement point planning method

Due to the complex surface profiles of blades, the current measurement point planning does not adequately capture the distinct zones, leading to low measurement efficiency and challenges in meeting online detection requirements. This study proposes a novel approach to blade zoning measurement point planning. First, it utilizes an improved centroid Voronoi extreme model and smoothness indicators to select seed points during the zoning process. Following that, feature tensor-driven regional clustering is carried out based on these seed points. Finally, measurement points are determined through adaptive sampling of profile deviations along the feature stacking axis within each zone. Simulation and real experiments demonstrate that the measurement points generated by this method effectively represent the blade surface's distinctive features, reducing the number of measurement points by 91.7% and significantly enhancing efficiency.

Numerical and experimental investigations in a cracked wind turbine blade using natural frequencies

In this study, the vibration behavior of a wind turbine blade under influence of the transverse crack was studied. A mathematical model based on free vibration of the blade with induced transverse crack was prepared to estimate the natural frequencies. Wind turbine blade was modeled considering double tapered beam with linearly varying width and thickness keeping constant aspect ratio. A Matlab program was constructed to obtain the frequencies from mathematical model. Modal analysis of wind turbine blade was performed simulating the induced transverse crack of different depths at various locations to estimate the natural frequencies using Fast Fourier Transform (FFT) analyzer. To verify the results obtained from mathematical model and modal analysis, finite element analysis was performed. Same methodology of FEA was implemented on another blade with different aspect ratio and blade profile to verify the obtained results. It was observed that the methods adopted are in correlation with one another depicting reduction in natural frequencies as crack approaches to the free end of blade as well as for increased depth of crack signifying reduction in stiffness of the blade.

Cooling Efficiency of a NACA4412 airfoil: Numerical application

The temperatures imposed on the blades of the first stages of turbines are generally very high; these expose the latter to harmful thermal effects, pushing manufacturers to continually improve techniques for cooling the blades. It is true that by increasing the temperature of the gases at the inlet of the turbines, we increase the efficiency, the performance of the machines, and we improve the power and fuel consumption with a significant reduction in polluting gases. Thus, the current general trend among manufacturers is to design machines that operate at increasingly high inlet temperatures. This has led, therefore, to the constant search for new materials with high thermal resistance and to constantly improve cooling techniques. This task is conditioned by a good and deep understanding of the phenomenon of heat transfer in turbine blades. This study examines the three dimension numerical simulation of the flow and heat exchange inside an internal cooling channel of a gas turbine blade with the profile of NACA 4412. This channel plays an important role in increasing heat exchange between the cooling air and the walls of the blade. In the turbulent regime, we have investigated the cooling of a profile blade NACA 4412 using forced convection (V = 200, 250, 300, 350, and 400 m / s). determined that the value of the cold air's speed increases with the intensity of the secondary flow inside the morning. Results found that, The flow dynamics and kinematics of fluid particles alter significantly when the cooling air speed is increased, and the cooling level is enhanced.