Parametric study of the performance of an impulse-type turbine with CFD

Abstract

A water wheel is an old machine, used ages ago to provide mechanical energy from hydraulic energy. First, ancient societies used this energy directly to grind in order to get flour and other products. Later, when generators existed, they were employed to provide electrical energy. Today, there are still some isolated communities in rural areas whose main source of energy comes from water wheels. There have been many studies about water wheels and the designs that provide a maximum available installed power. This thesis will do a research on an already existing water wheel located at the Hydraulics Laboratory of the Department of Mechanical Engineering and Industrial Construction of the University of Girona. The water wheel-type is a horizontal axle with a very simple design of the blades. Although these kind of water wheels are known of not being the best designs, they can be very useful in developing countries because of their simplicity. The purpose of the research is to determine the efficiency of this water wheel with simulation techniques and to obtain the geometrical conditions that may provide high enough efficiency values for being a feasible design as an element for producing electricity in isolated communities of developing countries with high resources of water. For doing so, we will simulate the water wheel employing the ANSYS 15.0 software package and choosing the commercial code CFX, which is a general purpose Computational Fluid Dynamics CFD model. After setting up the model, we will do a parametric study to optimize the efficiency of the water wheel. Firstly, a geometrical model of the water wheel will be designed to be as similar as possible to the actual water wheel, being a copy of the current water wheel of the laboratory. This includes every important piece of the water wheel and also the holes in the plates designed to lower the weight. We have to keep in mind, however, that a good geometrical design has to be accurate enough to reproduce the geometry of the real water wheel, but simple enough to avoid excessive un important details so we may have the results using reason able computational resources. Secondly, a study on the suitable mesh is required. Because of the limit of 512.000 elements in the educational version of ANSYS 15.0, we have to decide which part is more important in the contribution to the hydraulic power in order to apply a finer mesh there. The blades of the water wheel will receive the water jet impact and will provide the reaction forces that make the water wheel rotate. It is reasonable that a smaller mesh size at the blade will allow to obtain better results. Also the accuracy of reproducing the path of the water jet is important because the reaction forces come from the impact of the water. Once the discretization of the volume (i.e., the mesh) has been defined, the setup has been created. A parametric study that takes into account the rotation speed of the waterwheel will be carried out in order to determine at which point the power reach the maximum value. The formula of power is equal to the torque multiplied by the angular speed. If we fix the water wheel to a speed of zero (at rest), there will be a maximum of torque but a minimum of power. On the other hand, the water wheel turning at a constant speed equal to the speed of the incoming waterjet will not provide torque although it will give to the water wheel a maximum value of the rotational speed but, again, a minimum of power. These two points will be the starting and ending points of the power curve. After determining the maximum power as a function of the rotational speed, a second parameter will be modified. This parameter will be the position of the inlet with respect to the water wheel location, where a perpendicular impact to the water wheel is required. Afterwards, we can compare the results with the experimental data to see how accurate the simulations are and how much the efficiency is improved