Applied Ocean Research 30 (2008) 305310

Contents lists available at ScienceDirect

Applied Ocean Research

journal homepage: www.elsevier.com/locate/apor

Shape optimization of two-dimensional cavitators in supercavitating flows, using NSGA II algorithm

R. Shafaghat ,1 , S.M. Hosseinalipour 2 , N.M. Nouri 3 , I. Lashgari 4
Department of Mechanical Engineering, Iran University of Science and Technology, Iran

article

info

abstract
The reduction of energy consumption of high speed submersible bodies is an important challenge in hydrodynamic researches. In this paper, shape optimization of two-dimensional cavitators in supercavitating flows is studied. A two dimensional supercavitation potential flow passes a symmetric two dimensional cavitator, which is placed perpendicular to the flow in a channel of infinite length and immediately a cavity is formed behind the cavitator. This is because of the generation of a gas or vapor cavity between the body and the surrounding liquid due to the change in a high speed flow direction passing the cavitator. Drag force acting on this supercavitating body dictates the thrust requirements for the propulsion system, to maintain a required cavity at the operating speed. Therefore, any reduction in the drag force, by modifying the shape of the cavitator, will lead to decrease this force. This study concentrates on the optimization of two dimensional cavitators in order to decrease drag coefficient for a specified after body length and velocity in a potential flow. To achieve this goal a multi-objective optimization problem is defined to optimize cavitator shapes in supercavitating flow. The so-called NSGA II (Non-dominated Sorting Genetic Algorithm) algorithm is used as an optimization method. Design parameters and constraints are obtained according to supercavitating flow characteristics and cavitator modeling and objective functions are generated using Linear Regression Method. The obtained results are compared with other classic optimization methods, like the weighted sum method, for validation. 2009 Elsevier Ltd. All rights reserved.

Article history: Received 2 July 2008 Received in revised form 11 January 2009 Accepted 22 February 2009 Available online 2 May 2009 Keywords: Multi-objective optimization Supercavitation Two dimensional cavitator Drag coefficient Cavitation number

1. Introduction Supercavitation is the cavitation effects which create a large bubble of gas inside a liquid, allowing an object to travel at great speed through the liquid by being engulfed by the bubble. The creation of the cavity (i.e., the bubble) reduces the total drag on the object and this makes supercavitation an attractive phenomena. In recent years, supercavitation has attracted growing interest due to its potential for vehicle maneuvering and drag reduction. Many problems in cavitation (e.g., cavitating flows around hydrofoils and bodies of revolution, design of supercavitating foils, ventilated cavities, etc.) have been investigated in detail both experimentally and theoretically. However, there are still many more problems that require more careful research [1]. In the past, the hydrofoil propeller design philosophy was to avoid cavitation for a widespread range of operating conditions.

However, nowadays demands for underwater vehicle speeds and higher propeller efficiency have made this design philosophy practically impossible to apply. The efficiency of a non-cavitating high-speed propeller is relatively low, due to the required large blade area that leads to high frictional losses. Therefore the two dimensional supercavitating hydrofoils with controlled amounts of cavitation are utilized for designing propellers with a small blade area. A numerical optimization technique is applied to the design of two-dimensional supercavitating hydrofoil sections (Fig. 1). The drag coefficient of a cavitator is one of the most important characteristics of supercavitating flows. Usually the drag can be characterized as pressure drag and viscous drag. In the case of supercavitation, the pressure drag is considerably much higher than the viscous drag, since the body does not come into contact with the fluid [15]. In a special cavity, the length of the drag coefficient is a function of cavitator geometric parameters. In order to find the best cavitator shape in special conditions, we have to use an optimization process. 2. Literature review Early research on supercavitating flows was performed by Reichardt [1], who experimentally studied the axisymmetric supercavitating flows. Efros [2] employed conformal mapping techniques to investigate the supercavitating flow problems. Tulin [3] introduced the use of perturbation methods for examination of two

Corresponding author. E-mail address: rshafaghat@iust.ac.ir (R. Shafaghat).

1 Ph.D. Student. 2 Assistant Professor. 3 Assistant Professor. 4 MSc. Student. 0141-1187/$ see front matter 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.apor.2009.02.005

306

R. Shafaghat et al. / Applied Ocean Research 30 (2008) 305310

Fig. 1. Two-dimensional supercavitating hydrofoil sections.

Nomenclature Cd D d l P PC Pd P qc U x Drag Coefficient Drag Cavitator height Cavity length Half height of cavitator Cavity Pressure Pressure on the cavitator Far field pressure Velocity on the cavity Far field vlosity Control point Weight of each point Density Cavitation number Stream function

dimensional supercavitating flows. Cuthbert and Street [4] used sources and sinks along the axis of a slender axisymmetric body cavity system, along with a Riabouchinsky cavity closure model. They solved the problem for the unknown cavity shape, but they were successful only for few cases. Brennan [5] employed a relaxation method in a transformed velocity potential-stream function plane for analyzing axisymmetric cavitating flows behind a disk and a sphere between solid walls. Chou [6] extended the work of Cuthbert and Street [4] to solve axisymmetric supercavitating flows using slender body theory. Until the 1970s, the analytical methods were the most important ones to solve the supercavitating flow problems. Beginning in 1980, some numerical methods were also developed. Aitchison [7] used a method of variable finite elements to consider the flow past a disk in a tube of finite diameter and infinite length. Uhlman [8] used the surface singularity method to solve the fully nonlinear potential flow past a supercavitating flat-plate hydrofoil numerically. Hase [9] employed Interior source methods for modeling the planar and axisymmetric supercavitating flows. Verghese, Uhlman and Kirschner [10] used the boundary element method for numerical analysis of high speed bodies in partially cavitating axisymmetric flow. Shafaghat et al. [11] used the boundary element method for numerical analysis of two dimensional bounded supercavitating flows. The propulsion system provides the required thrust which is proportional to the drag which is exerted on the body. This act keeps the high speed motion of the body. Thus any decrease in the thrust by modifying the shape of cavitator is desirable. Kinnas et al. [12] and Mishima & Kinnas [13] studied the flow around the supercavitating hydrofoils and supercavitating

wings and obtained the optimized shape of hydrofoils, using Lagrange multiplier method. Mishima [14] presented his studies on cavitator modeling and hydrofoil optimization using a method of multipliers and a penalty parameter update scheme in both constrained and unconstrained optimization problems. Alyanak et al. [15] have designed the variable shape of cavitators. They have adjusted the cavitator parameters to obtain an optimized cavitator shape of a supercavitating torpedo. They introduced some nondimensional parameters in their cavitator modeling [16]. Choi [17] has investigated the cavitator shape optimization procedure using design sensitively analysis. He used a different method for the geometric definition of the cavitator in his work. It is noticeable that all researchers have used a type of gradient methods to optimize the cavitators or hydrofoils shape. These methods have been generally criticized for their problems with obtaining optimum points in uneven and uncontained objective functions. So a non-gradient method is considered in this paper. The geneticbase optimization methods utilize a sorting algorithm to obtain optimum points in domains of interest. In this study, a multiobjective genetic algorithm is selected to optimize the cavitator shape. It is confirmed by previous studies that the potential flow assumption is accurate enough for supercavitating flows analysis. Usually the main parameters in supercavitating flows are the geometry of cavity and cavitator and also the cavitation number. Having specified the objective functions and the design parameters, a multi-objective optimization procedure is used in order to minimize the drag coefficient and maximize the cavitation number. The required input data for this investigation are produced using a BEM supercavitating flow software (developed by the authors) [11] and the so called NSGA II [18,19] optimization algorithm. 3. Supercavitating flow modeling Supercavitation is a phenomenon in which a cavity is formed behind a moving body in the water (Fig. 2). As the body is enveloped inside the cavity and there is no direct contact between the body surface and the water, the the hydrodynamic drag is the only type of drag which has to be considered in such flows [17]. This is obtained by integration of the pressure distribution over the cavitator, as defined in Eq. (1). Cd = D
1 2

U2

2 d
0

d 2

Pd PC
1 2 2 U

dy.

(1)

The shape of the cavitators not only affects the cavity size but also determines the magnitude of the drag exerted on it. In this study

R. Shafaghat et al. / Applied Ocean Research 30 (2008) 305310

307

Fluid Cavity Body X2 b2

X3

z motion of body
Fig. 2. Supercavitating body in cavity.

X1 b1
Fig. 4. Cavitator geometrical modeling.

5. Shape optimization problem

H D

The main parameters in the shape optimization problem are defined as the following:
b

C A a B d
Fig. 3. Definition of the problem region [11].

Cd : Drag coefficient l/d: Ratio off cavity length to cavitator width : Cavitation number b1 , b2 and 2 : Cavitator geometric parameters.

the optimum cavitator shapes are developed for predetermined the cavity lengths. The governing equations and the appropriate boundary conditions for a potential flow across the cavitator and the supercavitation bubble are summarized as the following, (Fig. 3).

According to the definition of the cavitation number, is normally defined based on free-stream velocity and far field pressure.

P PC
1 2 2 U

2 =

2 2 + = 0 in ABCDEFA x2 y2
(2)

= 0 on AB, BC, CD = 0 on AF, DE n = 1 on EF = qc on CD. n

Based on the above equations, software is developed in which the cavity shape can be determined for a predetermined cavitator shape and supercavity bubble length using an iterative method by following the convergence criteria: 1. Impermeability in cavity boundary. 2. Constant pressure in cavity boundary. 4. Cavitator modeling A curve fit procedure being applied on a number of specific points for the cavitator geometry is used in order to define the complete geometry. The rational parabolic curve is employed in this study (Fig. 4).

However, considering a supercavitating object by specified cavity length to cavitator width ratio, the geometrical parameters of cavitators and the free stream physical conditions relate to each other. It means that the cavitation number and geometrical parameters are accompanied to demonstrate cavity relative length. Therefore, when this length regarding a fix, the cavitation number can be a function of geometrical parameters of cavitator. In order to guarantee the physical solutions, the constraints are to be specified exactly and correctly. According to the references [15,17] the best range of the l/d ratio is defined as the following: 5 l d

20.

(4)

The studied values for l/d ratios were 5, 10, 15 and 20 in this research. Based on the previous experiences, the following values were considered for geometric parameters as the necessary constraints in this study: 0 b1 1 0 b2 1 0 2 2. It should be noted that b1 and b2 are normalized values using the cavitator width as the normalization parameter. Beyond a specific value of b1 , the frictional loss on cavitator increases and therefore the total drag coefficient will increase too. Different sets of input parameters were defined and applied to the software in order to calculate the corresponding drag coefficient and cavitation number for each set. The objective functions were defined as follows: Cd = f1 (b1 , b2 , 2 ) (5)

(1 )2 x1 1 + 2(1 )x2 2 + 2 x3 3 x() = (3) (1 )2 1 + 2(1 )2 + 2 3 where x1 = (b1 , 0), x2 = (b1 , b2 ), x3 = (0, 1), 1 = 1 and 3 = 1. In this equation, x1 to x3 are the control points to describe quadratic curve, and 1 to 3 be the weight of each point. The design parameters are b1 , b2 and 2 is the weight value for the second point, x2 . If 2 increase, then the curve shape get closer to x2 , whereas if 2 decreases to zero, the curve becomes just a
straight line connecting x1 and x3 . Therefore the cavitator geometry is determined by three parameters, b1 , b2 and 2 .

= f2 (b1 , b2 , 2 ).

(6)

As the velocity and energy consumption decreases with increasing

308

R. Shafaghat et al. / Applied Ocean Research 30 (2008) 305310

the cavitation number, it is desirable to maximize the cavitation number. This causes the reduction of the drag coefficient. Finally the optimization problem is defined as follows: Min Cd = f1 (b1 , b2 , 2 ) Max = f2 (b1 , b2 , 2 ) 0 b1 1 0 b2 1 0 2 2 which is a multi-objective optimization problem. 6. Multi-objective optimization method (NSGA-II) Over the past decade, a number of multi-objective optimization algorithms have been developed. These algorithms have the ability to find Pareto-optimal solutions in one single run. These methods give a large number of alternative solutions lying on or near the Pareto-optimal ant which produce valuable and practical values. It should be noted that all researchers have used a type of gradient methods to optimize the cavitators or hydrofoils shape. These methods have been generally criticized for their problems in obtaining optimum points in uneven and uncontained objective functions. So a non-gradient method is considered in this paper. With respect to Eq. (7), the optimization problem is a multiobjective problem. Since the principal reason why a problem has a multi-objective formulation is because it is not possible to have a single solution which simultaneously optimizes all objectives, an algorithm that gives a large number of alternative solutions lying on or near the Pareto-optimal front is of great practical value. In contrast, the classic methods such as the weighted sum convert multi-objective problem to a single objective and then optimize it. These algorithms mostly offer a series of coefficients which should be selected by a designer in order to find a single solution in each run. Therefore each solution is dramatically dependent on the amount of these coefficients. In the comparison chart, each point of the results in the classical method is found separately due to the specified amount of m factor (Eq. (8)). We have proposed a computationally fast elitist multi-objective algorithm based on a non-dominated sorting approach. With the properties of a fast non-dominated sorting procedure, an elitist strategy, NSGA-II will have increasing attention and application in the near future. NSGA-II algorithms have the ability to find Paretooptimal solutions in one single run. In NSGA-II method, once the population initialized, the population is sorted based on non-domination into each front. The first front is completely non-dominant and set in the current population, and the second front is dominated by the individuals in the first front only and the front goes on like this. Each individual in each front is assigned rank (fitness) values, based on which front they belong to. In addition to fitness value, a new parameter called crowding distance is calculated for each individual. The crowding distance is a measure of how close an individual is to its neighbors. A large average crowding distance will result in better diversity in the population. Parents are selected from the population by using binary tournament selection based on the rank and crowding distance. An individual is selected in the rank which is lesser than the other, or if crowding distance is greater than the other. The selected population generates from crossover and mutation operators. The population with the current population and current offspring is sorted again based on non-domination and only the best N individuals are selected, where N is the population size. The selection is based on rank and on the crowding distance in the last front. Fig. 5 presents the results of applying the NSGA-II for each value of l/d ratio in our optimization problem. (7)

Fig. 5. Cd optimized curves for two dimensional cavitators.

Fig. 6. Comparison between NSGA-II & weighted sum optimization method for l/d = 10.

The results of the other classic optimization algorithms are considered as validation tools for the NSGA-II results. So one also considers the results which are obtained from weighted sum methods for validation. This classic method converts a multiobjective problem to a single objective one and optimizes it. Considering this, our optimization problem can be cast in the following form: Cd = f1 (b1 , b2 , 2 )

= f2 (b1 , b2 , 2 )
Min m f1 (1 m) f2 0 b1 1 0 b2 1 0 2 2. This problem for each l/d ratio is solved using different values of m between 0 and 1. Fig. 6 present a good agreement between NSGA-II and Weighted sum methods results for l/d = 10 ratio. 7. Optimized cavitator shapes In this study the pareto-optimal curves are used for producing the cavitators optimum shapes. Table 1 presents our study results for a high speed body moving with 33 m/s inside a 25 C water 10 m under the free level. As it is seen, the optimum cavitator shapes are tabulated for values of l/d and a specified value of . The obtained values of drag coefficients are also presented in the table for each set o data. (8)

R. Shafaghat et al. / Applied Ocean Research 30 (2008) 305310 Table 1 Optimized two dimensional cavitators data for = 0.36. Cavitation number = 0.36 L/d = 5 b1 1.00 b2 0.51 Cd = 0.5472 w2 1.86 L/d = 10 b1 0.69 b2 0.58 Cd = 0.6401 w2 1.99 L/d = 15 b1 0.42 b2 0.66 Cd = 0.7597 w2 1.98 L/d = 20 b1 0.36 b2 0.77 Cd = 0.8189

309

w2 1.63

Fig. 9. Actual cavitator shape in a fixed cavitation number. Fig. 7. Optimized two dimensional cavitators shape for = 0.36.

Fig. 10. Optimized two dimensional cavitators shape for Cd = 0.7.

Fig. 8. Cavity detachment point for = 0.36.

Fig. 7 depicts the half upper part of cavitator shape, based on the obtained geometric parameters. It should be noted that the cavities detachment points specify the actual cavitator shape. According to Brillouin condition [20,21] for cavity detachment point, cavities start points are presented in Fig. 8. Using this information, the final cavitator shape is presented in Fig. 9. Table 2 presents the results for the second case in which the drag coefficient is considered as a fixed value. It should be noted that the drag coefficient dictates the propulsion force. Therefore, the cavitator shape is specified for each l/d ratio and its corresponding cavitation number using Pareto optimal curves. Fig. 10 depicts the half upper part of cavitator shape, based on the obtained geometric parameters. The final cavitator shape (using Brillouin detachment condition) are presented in Fig. 11.

Fig. 11. Actual cavitators shape for Cd = 0.7.

8. Conclusions The results of a shape optimization problem for supercavitating flows are presented in this paper using the NSGA-II optimization technique. The main variables are the cavitators geometric

310

R. Shafaghat et al. / Applied Ocean Research 30 (2008) 305310

Table 2 Optimized two dimensional cavitators data for CD = 0.7. Drag Coefficient = 0.7 L/d = 5 b1 0.812 b2 w2 L/d = 10 b1 0.615 b2 0.595 Sigma = 0.3855 w2 2 L/d = 15 b1 0.564 b2 w2 L/d = 20 b1 0.594 b2 0.637 Sigma = 0.3215 w2 1.554

0.538 1.703 Sigma = 0.4618

0.652 1.852 Sigma = 0.3366

parameters and the supercavitating flow characteristics, namely and Cd . The regression method was used to produce the final objective functions for the problem in hand. Optimization of the cavitator shape to minimize drag, which is the main goal of this study, was then carried out using a multi-objective optimization technique, in which the drag coefficient was minimized while the cavitation number was maximized in a single optimization routine. The cavitator optimization was then carried out for a number of cavity lengths to cavitator ratios, based on the previous investigation data. Optimization results showed that, for a specified cavitation number, as the required cavity length becomes longer, the drag coefficient increases and the optimum shape of cavitator changes from a long flat nose shape to a short smooth one. It also showed that for a specified drag coefficient, as the required cavity length becomes longer, the cavitation number decreases and the optimum shape of cavitator changes from a long flat nose shape to a short smooth one. The authors have extended this method for a three-dimensional supercavitating flow. According to the defined problem, the concept and material of two-dimensional cavitator optimization are used to achieve optimized curves for axisymmetric cavitators. References
[1] Reichardt H. The physical laws governing the cavitation bubbles produced behind solids of revolution in a fluid flow. Rep. UM 6628. Gottingen: The Kaiser Wilhelm Institute for Hydrodynamic Research; 1945. [2] Efros DA. Hydrodynamic theory of two-dimensional flow with cavitation. Dokl Akad Nauk SSSR 1946;51:26770. [3] Tulin MP. Supercavitating flowsSmall perturbation theory. J Ship Res 1964; 7:3. [4] Cuthbert J, Street R. An approximate theory for supercavitating flow about slender bodies of revolution. LMSC Report, TM81-73/39. Sunnyvale (CA): Lockheed Missiles and Space Co.; 1964. [5] Brennen CA. Numerical solution of axisymmetric cavity flows. J Fluid Mech 1969;37:67188. 4.

[6] Chou YS. Axisymmetric cavity flows past slender bodies of revolution. J Hydronautics 1974;8:1. [7] Aitchison JM. The numerical solution of planar and axisymmetric cavitational flow problems. Comput & Fluids 1984;12(1):5565. [8] Uhlman JS. The surface singularity or boundary integral method applied to supercavitating hydrofoils. J Ship Res 1989;33:1. [9] Hase PM. Interior source methods for planar and axisymmetric supercavitating flows. Thesis submitted for the degree of Doctor of Philosophy, May 2003. [10] Verghese AN, Uhlman JS, Kirschner IN. Numerical analysis of highSpeed bodies in partially cavitation axisymmetric flow, transactions of ASME. J Eng 2005;127:4154. [11] Shafaghat R, Hosseinalipour SM, Shariatifard A. Numerical analysis of a two dimensional bounded supercavitation flow. 15th Annual Conferences of the CFD Society of Canada; 2007. [12] Kinnas SA, Mishima S, Savineau C. Application of optimization techniques to the design of cavitating hydrofoils and wings. Department of Ocean Engineering Massachusetts Institute of Technology, Present at the International Symposium on Cavitation Deauville, France, 25 May; 1995. [13] Mishima Sh, Kinnas SA. A numerical optimization technique applied to the design of two-dimensional cavitating hydrofoil sections. J Ship Res 1995; September. [14] Mishima Shigenori. Design of cavitating propeller blades in non-uniform flow by numerical optimization. For the Degree of Doctor of Philosophy at the Massachusetts Institute of Technology, June 1996. [15] Alyanak Edward, Venkayya Vipperla, Grandhi Ramana, Penmetsa Ravi. Variable shape cavitator design for a supercavitating torpedo, department of mechanical and material engineering. Dayton (U S): Wright State University, OH 45345. [16] Alyanak Edward, Grandhi Ramana, Penmetsa Ravi. Optimum design of a supercavitatiting torpedo considering overall size, shape, and structural configuration, Dayton,OH 45345 (United State): Department of mechanical and material engineering, Wright State University; 22 July 2005. [17] Choi JH, Penmetsa RC, Grandhi RV. Shape optimization of the cavitator for a supercavitating torpedo. Struct Multidisk Optim 29, 159167. Aerospace and Mechanical Engineering, Hankuk Aviation University; 2005. [18] Deb Kalyanmoy, Agrawal Samir, Pratap Amrit, Meyarivan T. A fast elitist nondominated sorting genetic algorithm for multi-objective optimization: NSGAII. India: Indian Institute of Technology Kanpur, PIN 208 016; 2000. [19] Seshadri Aravind. A fast elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation 2004;(April). [20] Knapp RT, Dailly JW, Hammitt FG. Cavitation. McGrow-Hill Book Company; 1970. [21] Armstrong AH. Abrupt and smooth separation in plane and axisymmetric flow. Memor Arm Res Est 1953; G.B. no. 22/63.