Introduction

The velocity of water in a river is a clean and constant source of energy. To use this without having to build a large infrastructure, such as that needed in large hydroelectric plants, various types of turbines submerged in water currents, called hydrokinetic turbines (THCs), have been developed. These turbines do not depend on high altitudes or waterfalls, which makes them a low-cost technology and reduces implementation time, in addition to having a minimal environmental impact (^{Chica, Perez, Rubio-Clemente, & Agudelo, 2015}; ^{Filho, Souza, Rossi, Barros, & Silva, 2010}; ^{Gaden & Bibeau, 2010}; ^{Yuce & Muratoglu, 2014}). These turbines extract kinetic energy from marine currents, rivers, and artificial canals, among other sources, and convert it into mechanical power without interrupting the natural flow of the water (^{Badea, Pricop, & Bobonea, 2014}). However, these devices present a great disadvantage with respect to the low energy density obtained in comparison with conventional hydroelectric power plants, suggesting that their economic feasibility must be studied in detail (^{Gaden & Bibeau, 2010}; ^{Khan, Bhuyan, Iqbal, & Quaicoe, 2009}). THCs can be characterized by the orientation of their rotational axis with respect to the direction of water flow, classifying them as axial flow and cross-flow, or transversal turbines. Axial flow THCs have their rotational axis parallel to the direction of the current, using propeller-type impellers. Several layouts of this type of turbines are shown in the Figure 1. The tilted shaft turbine (Figure 1a) is most often used in small rivers, while the other arrangements (Figures 1b, Figure 1c, and Figure 1 d) are mainly used for ocean energy extraction ((^{Vermaak, Kusakana, & Koko, 2014}).

Figure 1 Hydrokinetic axial flow turbines: a) inclined axis, b) rigid mooring, c) non-submerged generator, and d) submerged generator. Source: ^{Vermaak et al. (2014}). 

Meanwhile, cross-flow or transversal hydrokinetic turbines (THCFC or THCFT) have their rotation axis perpendicular to the direction of the water flow and can operate regardless of its direction. These turbines can be classified into two types. One has horizontal axis turbines in a transverse arrangement (Figure 2a), which operate based on the drag force and have a poorer performance than the vertical axis. The other type, consisting of vertical axis turbines (Figure 2b, Figure 2c, Figure 2d, Figure 2e, and Figure 2f) operates based on the lift force, and is widely used in hydroelectric applications, thanks to the advantages offered by the horizontal shaft turbines in a transversal layout. However, their hydraulic design and behavior are more complex ((^{Zanette, Imbault, & Tourabi, 2010}).

Figure 2 Cross-flow hydrokinetic turbines: a) horizontal axis, b) squirrel cage Darrieus, c) H-Darrieus, d) Darrieus, e) Gorlov, and f) Savonius. Source: ^{Vermaak et al. (2014}). 

The principle of operation of the THCFC or THCFT consists of the transformation of the kinetic energy of the water contained in a natural water source into the turbine's rotational kinetic energy. This occurs due to the water current that flows through the blades, which are attached to a rotor. With some of the geometries of vertical axis turbines, such as those shown in Figure 2b, Figure 2c, Figure 2d, and Figure 2e, the water comes into contact with the blades. This takes place in two stages: first, the fluid enters from the outside towards the interior, and second, water flows from the inside to the outside of the turbine, thereby transferring its energy (^{Okot, 2013}). This is different than horizontal axis hydrokinetic turbines, which transfer the kinetic energy of the fluid with only one stage, and generate higher speeds of rotation than cross-flow.

The energy contained in a stream of water is directly proportional to the density of the fluid, the area of the cross section, and the velocity of the fluid. Therefore, when comparing THCs with speeds from 1.75 m/s to 2.25 m/s, and wind turbines with speeds between 11 m/s and 13 m/s, there is a great difference in the power generation potential, since the density of water is 832 times greater than that of air under normal temperature and pressure conditions. It is important to mention that the fluid that passes through the turbine does not deliver all its energy. In fact, a power coefficient has been defined which indicates the amount of kinetic energy that can be extracted from the flow and converted into mechanical energy in the axis of the turbine, with a maximum value of 59.3%, which is known as the Betz limit (^{Koko, Kusakana, & Vermaak, 2015}; ^{Shahsavarifard, Bibeau, & Birjandi, 2013}; ^{Vermaak et al., 2014}). Subsequently, an extension of this theory is presented, known as the Lanchester-Betz limit, which indicates that not all the available energy can be captured by the turbine, because this would mean a complete obstruction of the fluid (^{Ruopp, Ruprecht, Riedelbauch, Arnaud, & Hamad, 2014}). The total electrical power that is generated is affected not only by this coefficient but also by the efficiencies of the transmission mechanism and the generator, the physical properties of the fluid (density), the area swept by the turbine blades, and the operating conditions (flow) (^{Gaden & Bibeau, 2010}; ^{Guney, 2011}). Two other features that makes THCs a more attractive alternative are its modularity and expandability. They can be arranged in arrays of multiple units in order to extract more energy from the water streams (^{Zanette et al., 2010}).

Hydrokinetic turbines

Although various technologies use renewable energy to generate electricity, some cannot be used or do not have the necessary conditions to obtain the optimal efficiency needed to maintain a good cost-benefit ratio, thus becoming unsustainable technologies. This is often found in areas with special environmental or geographical characteristics, where high rainfall occurs with long periods of cloudy sky, which along with high humidity and rapid growth of fungi and vegetation generates premature failures in solar energy panels. In addition, densely wooded areas do not have sufficient air currents to supply a wind turbine, and water sources may not be high enough to feed conventional electric microgeneration systems. For this reason, hydrokinetic turbines emerge as an alternative capable of satisfying the energy demand in these areas, if they are provided with a water source (^{Anyi & Kirke, 2015}).

Currently, very few studies have been carried out on THCs, since it is a technology that is in its initial stage and needs to be investigated in greater depth (^{Yuce & Muratoglu, 2014}). The lack of correlations characterizing this type of turbomachinery is evident in the literature, which is why most studies on THCs have been developed empirically, especially at the design stage, where the support profiles used in most of these projects have been adapted from results obtained by tests performed on wind turbines.

Design of horizontal axis hydrokinetic turbines

THCs operate under the same principles of wind turbines, sharing similar design philosophies (^{Anyi & Kirke, 2011}; ^{Chica et al., 2015}; ^{Day et al., 2015}; ^{Güney & Kaygusuz, 2010}; ^{Yuce & Muratoglu, 2014}). The design of a THCEH starts with determining the dimensions of the rotor, which is done using Equation 1, based on the quantity of the power output P required or expected from the turbine, which is given in terms of the density, ρ, and velocity, V, of the fluid, the area of the cross section, A swept by the blade (disc model), the power coefficient, C _p , and the efficiency of the transmission train, η (^{Anyi & Kirke, 2011}; ^{Chica et al., 2015}; ^{Vermaak et al., 2014}), where the term C _p η should be equal to 21% (^{Anyi & Kirke, 2011}):

An important parameter in the design of the THCEH is strength, which is the ratio between the pitch and the rope of the blades, and it should be around 30%, in order to have a good starting torque and a high relative speed (^{Anyi & Kirke, 2010}). Another important factor is the angle of torsion of the blade from the tip to the base, since a study by ^{Lanzafame and Messina (2009)} on designs of blades for wind turbines showed that a twisted blade performs better than one that is totally straight. Other factors of interest exist, such as the selection of the hydrodynamic profile to be used. Several aspects must be taken into account when designing a THC, for example, the type of turbine to be used. The performance coefficients for the different existing turbines must be taken into account, which are commonly given as a function of the tip speed ratio (TSR), defined as "the relationship of the tangential velocity at the tip of the blade to the reference velocity of free flow" (^{Chica et al., 2015}). The TSR is represented according to Equation (2), where R is the radius swept by the blade of the turbine, ω the speed of the blade’s rotation, and V is the velocity of the fluid:

Figure 3 shows the TSR vs power coefficient diagram, where the cross flow turbines work at low speeds and have low power coefficients compared to the axial turbines (^{Guney, 2011}). This fact must also be considered when selecting the type of turbine. Another important aspect to keep in mind is the variation in the current velocity profile, for the purpose of determining the point where the system should be installed in order to obtain better efficiency. Figure 4 schematically illustrates how velocity varies with depth, observing that in the background the velocity is low due to friction with the bottom of the channel, and it increases to a maximum value, then decreases as it reaches the surface due to friction with the air. The velocity also varies laterally along the width of the channel. This was analyzed by ^{Gunawan et al. (2012)}, who determined that the maximum velocity is near the center of the current. ^{Neary, Gunawan and Sale (2013)} were able to demonstrate that turbulence can increase the hydraulic efficiency of the turbine by approximately 12%, so to take advantage of this additional energy, the effect of this turbulence on the structural design of the system should be considered.

Figure 3 TSR vs. C _p diagram with different turbines. Source: adapted from ^{Guney (2011}). 

Figure 4 Schematic representation of the relationship between depth and change in current velocity. Source: ^{Gunawan et al. (2012}). 

Machines used to increase the efficiency of hydrocintic turbines

^{Liu and Packey (2014)} proposed using THCs directly at the output of already established turbines at conventional hydroelectric power stations, or at sites neighboring these power plants, which is called the Combined Electric Power Cycle System (CCHS). They analyzed the possible changes, advantages, and potential benefits that this would entail, and reported that the CCHS concepts have been used and demonstrated in two projects carried out in the USA. The authors concluded that despite being an attractive technology, its viability requires a cost-benefit analysis.

Perhaps the biggest obstacle faced by THCs, in terms of their implementation in rivers, are the suspended solids from human activities. This problem has been largely avoided by installing these devices in areas where there is not too much dirt, which significantly limits its location, a fundamental factor in distributed generation (^{Anyi & Kirke, 2015)}. In order to solve this problem, ^{Anyi and Kirke (2015)} developed a THCEH capable of swinging its blades back and forth in its plane of rotation to enable the slipping of residues, and which has an internal ring that prevents grass or other elements from winding up around the axis, obtaining as a result a turbine that is practically free of weeds. This turbine operates normally when it is in a relatively clean environment (Figure 5a), can adapt to the presence and interference of residues, allows the radial displacement of each of its blades (Figure 5b), which are capable of deformation or inclination downstream of the turbine due to axial loads (Figure 5c). Thus, it acts as a cleaning and protection mechanism to allow the passage of debris, enabling the vanes to return to their initial position.

Figure 5 Hydrokinetic turbine operations: a) normal operation, b) capacity of the blade to move angularly on its axis of rotation, and c) ability to tilt the blades downstream of the turbine (^{Anyi & Kirke, 2015}). 

Conclusions

The studies carried out on hydrokinetic turbines show that this is a young technology with a large field of application, which can be used to supply electricity to areas that have nearby water sources without the need for falls or water heights, making this an alternative requiring little financial investment, and which is affordable and easy for the population to maintain, who are generally in isolated areas and have little income and an adequate quality of life.

The field of research on hydrokinetic turbines is broad, since the studies carried out have been based on the adaptation of analyses and results from wind turbines, and the technology is in a trial-and-error stage. There is still a large number of hydrodynamic profiles to be analyzed, in addition to the implementation of various improvements and modifications made to wind mechanisms in order to increase the efficiency of hydrokinetic turbines.

There is evidence of growing confidence in the use of computational simulations to analyze the hydrodynamics and structures of hydrokinetic turbines, enabling them to be more easily optimized, greatly reducing manufacturing costs.

There is a great tendency to use diffusers in order to improve the efficiency of hydrokinetic turbines. And one of the major problems with this technology is the debris, weeds, and other foreign elements present in the rivers. These are two interesting starting points that can be addressed by future studies on hydrokinetic turbines.