Design Tools for the Performance Improvement of a 76 MW Francis Turbine Runner

2009 ◽  
Author(s):  
Jose´ Manuel Franco-Nava ◽  
Oscar Dorantes-Go´mez ◽  
Erik Rosado-Tamariz ◽  
Jose´ Manuel Ferna´ndez-Da´vila ◽  
Reynaldo Rangel-Espinosa
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Finite Element ◽  
Performance Improvement ◽  
Three Dimensional ◽  
Francis Turbine ◽  
Design Tools ◽  
Turbine Runner ◽  
Induced Stresses ◽  
Fluid Behaviour

Application of two mayor design tools, Finite Element Analysis (FEA) and Computational Fluid Dynamics (CFD), for the performance improvement of a 76 MW Francis turbine runner is presented. In order to improve the performance of the runner, not only a CFD based optimization for the runner but also its structural integrity evaluation was carried out. In this paper, a number of analyses included within the design tools-based runner optimization process are presented. Initially, a reference condition for the fluid behaviour through turbine components was carried out by means of the computation of fluid conditions through the spiral case and stays vanes, followed by CFD-based fluid behaviour for the wicket so as to include the flow effects induced by these components in the final CFD analysis for the runner. All CFD computations were generated within the three dimensional Navier-Stoke commercial turbomachinery oriented CFD code FINE™/Turbo from NUMECA. The whole hydraulic turbine performance was then compared against actual data from a medium-head Francis type hydro turbine (76 MW). Then, CFD-based flow induced stresses in the turbine runner were computed by using a three dimensional finite element model built within the FEA commercial code ANSYS. Appropriate boundary conditions were set in order to obtain the results due to the different type loads (pressure and centrifugal force). The FEM model was able to capture the pressure gradients on the blade surfaces obtained from the CFD results. Improvement of efficiency and power for the runner was computed by using a parametric model built within 3D CFD code integrated environment FINETM/Design3D from NUMECA which combines genetic algorithms and a trained artificial neural network. During the optimization process the artificial neural network is trained with a database of geometries and their respective CFD computations in order to determine the optimum geometry for a given objective function. The optimisation process and the trend curve of the optimization or design cycle that included 29 parameters (corresponding to the control points of runner blade primary sections) which could vary during the process is presented. Finally, the flow induced stresses of the optimized Francis turbine runner was computed so as to evaluate the final blade geometry modifications related to the efficiency and power improvement.

Flow Induced Stresses in a Francis Turbine Runner Using Computer-Based Design Tools

2009 ◽  
Author(s):  
Jose´ Manuel Franco-Nava ◽  
Oscar Dorantes-Go´mez ◽  
Erik Rosado-Tamariz ◽  
Jose´ Manuel Ferna´ndez-Da´vila ◽  
Reynaldo Rangel-Espinosa
Keyword(s):  
Finite Element Analysis ◽  
Finite Element ◽  
Three Dimensional ◽  
Francis Turbine ◽  
Computational Domain ◽  
Element Analysis ◽  
Coordinate Measurement ◽  
The Finite Element Analysis ◽  
Turbine Runner ◽  
Induced Stresses

The stress analysis of the runner due to different loading is one of the most important tools that contribute its structural integrity evaluation. Finite element method has shown to be a strong numerical technique to provide good engineering accuracy. In this paper, the flow induced stresses in a Francis turbine runner is presented by using the finite element analysis. The runner geometry considered within the computational domain was modelled by using a three-dimensional laser triangulation scanner coupled with a portable coordinate measurement system. The runner geometry was generated by a number of 3D sub models, one for each of the main components of the runner, crown, band and a blade. In order to obtain a blade geometry a portable coordinate measurement system based on optical digitalization technology (scanner technology) was used. Because of symmetry, only a section of the runner domain was used for the finite element analysis. The runner was modeled with twenty-node solid elements. Loads due to pressure on the blade were derived from CFD computations for the runner at different power conditions (100%, 85% and 75%) for a medium head hydro power plant. CFD computations were carried out using the Finite Volume Method implement within FINE™/Turbo by NUMECA. The turbulence mathematical model used for the CFD computation was the Sparlart-Allmaras. The mesh of the turbine runner included different computational domains. For the runner blades the computational domain (mesh block) was defined in order to capture the complete blade row. All mesh blocks were structured hexahedral. Centrifugal force based on the rotational speed was considered. Also, a combined type loading analysis was computed including both pressure and centrifugal force. Appropriate boundary conditions were set in order to obtain the results due to the different type of analysis. The number of finite elements included in the FEM model was able to capture the pressure gradients on the blade surfaces obtained from the CFD results, which were investigated by application of a three dimensional Navier-Stoke commercial turbomachinery oriented CFD code. Analysis of the flow through the spiral case and stay vanes was carried out so as to include appropriate flow effects induced by these components and boundary conditions at the inlet of the wicket. A CFD analysis for the wicket and runner was carried out to generate the so called CFD reference solution. The analysis presented in this paper represents an initial characterization in order to increase understanding about combined loads acting on blades and to establish a reference state of stresses further comparison after refurbishments or optimization of the runner blades for a medium head hydroelectric power station.

Numerical Simulation of Francis Turbine

2014 ◽  
Author(s):  
Raza A. Saeed
Keyword(s):  
Finite Element ◽  
Fluid Flow ◽  
Water Pressure ◽  
Three Dimensional ◽  
Element Model ◽  
Francis Turbine ◽  
Three Dimensional Model ◽  
Element Analysis ◽  
Turbine Runner ◽  
Flow Through

This paper presents the results of modelling of the complete three-dimensional fluid flow through the spiral casing, stay vanes, guide vanes, and then through the Francis turbine runner to the draft tube of the Derbendikhan power station. To investigate the flow in the Francis turbine and also to compute stress distribution in the runner blades, a three-dimensional model was prepared according to specifications provided. The two topics discussed in this study are: (i) the simulation of the 3D fluid flow through the inter blade channels for the Francis turbine runner by using Computational Fluid Dynamics (CFD) and, (ii) the simulation of the stress analysis of the turbine runner by using Finite Element Analysis (FEA). In this study, the water pressure obtained from the CFD analysis for different boundary conditions are incorporated into a Finite Element model to calculate stress distributions in the runner.

scholarly journals Prediction of Stress in Power Transformer Winding Conductors Using Artificial Neural Networks: Hyperparameter Analysis

Energies ◽  
2021 ◽  
Vol 14 (14) ◽  
pp. 4242
Author(s):  
Fausto Valencia ◽  
Hugo Arcos ◽  
Franklin Quilumba
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Finite Element ◽  
Power Transformer ◽  
Network Models ◽  
Activation Function ◽  
Neural Network Models ◽  
Transformer Winding ◽  
Artificial Neural ◽  
Femm Software

The purpose of this research is the evaluation of artificial neural network models in the prediction of stresses in a 400 MVA power transformer winding conductor caused by the circulation of fault currents. The models were compared considering the training, validation, and test data errors’ behavior. Different combinations of hyperparameters were analyzed based on the variation of architectures, optimizers, and activation functions. The data for the process was created from finite element simulations performed in the FEMM software. The design of the Artificial Neural Network was performed using the Keras framework. As a result, a model with one hidden layer was the best suited architecture for the problem at hand, with the optimizer Adam and the activation function ReLU. The final Artificial Neural Network model predictions were compared with the Finite Element Method results, showing good agreement but with a much shorter solution time.

scholarly journals Artificial Neural Network-Based Method for Seismic Analysis of Concrete-Filled Steel Tube Arch Bridges

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Zhen Liu ◽  
Shibo Zhang
Keyword(s):  
Neural Network ◽  
Finite Element Method ◽  
Artificial Neural Network ◽  
Finite Element ◽  
Seismic Analysis ◽  
Steel Tube ◽  
Arch Bridge ◽  
Concrete Filled Steel Tube ◽  
Cfst Arch Bridge ◽  
Artificial Neural

Seismic analysis of concrete-filled steel tube (CFST) arch bridge based on finite element method is a time-consuming work. Especially when uncertainty of material and structural parameters are involved, the computational requirements may exceed the computational power of high performance computers. In this paper, a seismic analysis method of CFST arch bridge based on artificial neural network is presented. The ANN is trained by these seismic damage and corresponding sample parameters based on finite element analysis. In order to obtain more efficient training samples, a uniform design method is used to select sample parameters. By comparing the damage probabilities under different seismic intensities, it is found that the damage probabilities of the neural network method and the finite element method are basically the same. The method based on ANN can save a lot of computing time.

Sign in / Sign up

Export Citation Format

Share Document