International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

RESEARCH ARTICLE OPEN ACCESS

Design Optimization of Planar Mechanisms

Dr. Kailash Chaudhary
Department of Mechanical Engineering ,Raj Engineering College Jodhpur, India

Abstract:
This paper presents an optimization technique to dynamically balance the planar five-bar
mechanisms in which the shaking force and shaking moment are minimized using the genetic algorithm
(GA). A dynamically equivalent system of point-masses that represents each rigid link of a mechanism is
developed to represent link’s inertial properties. The shaking force and shaking moment are then expressed
in terms of the point-mass parameters which are taken as the design variables. These design variables are
brought into the optimization scheme to reduce the shaking force and shaking moment. This formulates the
objective function which optimizes the mass distribution of each link. The balancing problem is formulated
as a multi-objective optimization problem and multiple optimal solutions are created as a Pareto front by
using the genetic algorithm. The masses and inertias of the optimized links are computed from the
optimized design variables. The effectiveness of the proposed methodology is shown by applying it to a
problem of five-bar planar mechanism available in the literature.

Keywords — Dynamic balancing, Shaking force and shaking moment, Equimomental system,
Optimization, Genetic algorithm.

1. INTRODUCTION
An unbalanced mechanism running at high speed transmits forces and moments to the ground
known as shaking forces and shaking moments. These forces and moments are vector sum of
the inertia forces and moments of all the moving links. They adversely affect the dynamic
performance of the mechanism. Several techniques are presented in the literature for reducing
these shaking forces and shaking moments due to inertia. The complete force balancing can
be achieved by making the mass center of moving links of a mechanism stationary [1]. This
is achieved either by mass redistribution or by adding counterweights to the moving links.
This methodology was extended for the mechanisms having prismatic joints under certain
conditions [2, 3]. Force balancing and trajectory tracking is achieved in a five-bar real-time
controllable mechanism using adjusting kinematics parameter approach [4].
The complete force balancing increases other dynamic performance characteristics such as
shaking moment, driving torque and bearing forces in joints [5]. Therefore, along with the
full force balancing, several methods proposed in the literature to balance the shaking
moment [6, 7]. The complete force and moment balancing is achieved by adding duplicate
mechanism, inertia or disk counterweights [8-10]. However, this method is not recommended
due to complexity and practical reasons.
Several trade-off methods were developed to minimize different dynamic quantities
simultaneously [11, 12]. As the shaking force and shaking moment depend on link masses,
their locations of mass centers and moment of inertias, these trade-off methods find the
optimal distribution of the link masses [13].
The conventional optimization methods like gradient based search method is used to
optimally balance the planar mechanisms [14,15] and to analyse the sensitivity of shaking

ISSN: 2395-1303 http://www.ijetjournal.org Page 51

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

force and shaking moment to the design variables [16]. Optimum force balancing is achieved
for a five-bar mechanism using natural orthogonal complement dynamic modeling [17]. The
shaking moment is minimized in five-bar manipulator through constrained nonlinear
optimisation problem in which shaking force elimination is presented as the balancing
constraints [18]. The conventional optimization methods require an initial guess point to start
searching the optimum solution and likely to produce local optimum solution close to the
start point.
The evolutionary optimization techniques like particle swarm optimization (PSO) and genetic
algorithm (GA) can be applied to minimize multi-objective functions subject to some design
constraints [19, 20]. The mixed mass redistribution method using genetic algorithm is applied
for reducing shaking force and shaking moment in mechanisms [21].
In this paper, the formulation of optimization problem is simplified by modelling the rigid
links of mechanism as dynamically equivalent system of point-masses, known as
equimomental system [22, 23]. The balancing problem is formulated as a multi-objective
optimization problem and solved using genetic algorithm. This algorithm doesn’t require a
start point and searches the solution in the entire design space. Therefore, it produces the
global optimum solution for the optimization problem. Also, for a multi-objective
optimization problem, it produces several solutions which are all pareto optimum. Any
solution among these can be chosen as per the specific requirement.
The structure of this paper is as follows. Section 2 presents the equations of motion for rigid
body and equimomental point-masses. Problem of minimizing shaking force and shaking
moment for a five-bar mechanism is formulated in Section 3. A numerical example is
solved using the proposed method and its results are presented in Section 4. Finally,
conclusions are given in Section 5.

2. EQUIMOMENTAL SYSTEM OF POINT-MASSES

In this section, the concept of equimomental system of point-masses is discussed and the
dynamic equation of motion for a rigid body is rewritten in terms of the point-masses.

2.1. Equations of motion of rigid body

The links of a mechanism can be modelled as rigid bodies for simplifying the kinematic and
dynamic analyses. Consider an ith rigid link having motion in XY plane for which a local
frame, Xi Yi, is fixed at Oi on the body. The Newton-Euler (NE) equations of motion for this
link in the fixed inertial frame, OXY, are written as [11]:

M i t& i + C i t i = w i . (1)

ISSN: 2395-1303 http://www.ijetjournal.org Page 52

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

Oi+1

Ci ai
mi
Yi θi Xi
di
αi

Oi
Y

X
O

Figure 1. The ith rigid link moving in XY plane

In Eq. (1), 3 vectors, t i , t& i and w i are twist, twist-rate, and wrench vectors of the ith link with
respect to Oi, respectively, i.e.,

ωi  ω& i  ni 

t i =  ; t& i =   and w i =   (2)
vi   v& i  fi 

where, ω i and v i are the scalar angular velocity about the axis perpendicular to the plane of
motion and the 2-vector of linear velocity of the origin Oi, respectively.
Accordingly, ω& i and v& i are time derivatives of ω i and v i , respectively. Also, the scalar, n i , and
the 2-vector, f i , are the resultant moment about Oi and the resultant force at Oi, respectively.
In Eq. (1), the 3×3 matrices, Mi and Ci are defined as:

 Ii − mi d i sin(θ i + α i ) mi d i cos(θ i + α i )

M i =  − mi d i sin(θ i + α i ) mi 0 ;

 m i d i cos(θ i + α i ) 0 mi 
(3)
 0 0 0

C i = − ωi mi d i cos(θ i + α i ) 0 0
 − ωi mi d i sin(θ i +α i ) 0 0

Now, the points on the link, Oi and Oi+1 are defined at the joints connecting preceding and
succeeding links. The body fixed frame, Oi Xi Yi, is then defined in such a way that the axis Xi
is aligned from Oi to Oi+1. The shortest distance between Oi and Oi+1 is defined as link length.
The parameters di and θi are polar coordinates of the mass center as shown in Fig. 1.

2.2. Modified equations of motion for equimomental system of point-masses

To formulate an optimization problem to minimize shaking force and shaking moment, the
rigid links are modeled as dynamically equivalent systems of point-masses referred to
equimomental systems. The rigid link and the system of point-masses will be dynamically
equivalent (equimomental) if they have same mass, same center of mass and same inertia

ISSN: 2395-1303 http://www.ijetjournal.org Page 53

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

tensor with respect to same coordinate frame [22]. Hence, a set of dynamically equivalent
system of rigidly connected n point-masses, mij, located at lij, θij, as shown in Fig. 2 must
satisfy the following conditions:

∑m j
ij = mi (4)

∑m l
j
ij ij cos(θ ij + αi ) = mi d i cos(θ i + α i ) (5)

∑ m l sin(θ
j
ij ij ij + αi ) = mi d i sin(θ i + α i ) (6)

∑m l j
2
ij ij = Ii (7)

Oi+1

mi1 ai
Yi Xi
li1 θi2
θi1 lij mij

Oi
θij
Y li2
mi2

X
O

Figure 2. Equimomental system of point-masses for ith link

where mi and Ii are the mass of the ith link and its mass moment of inertia about Oi. The first
subscript i denotes the link number, and the second subscript j represents the point-mass. The
NE equations of motion, Eq. (1), are now rewritten for the equimomental system of point-
masses using equimomental conditions, Eqs. (4)-(7). It can be shown that the form, Eq. (1),
does not change except the elements of matrices, Mi and Ci, which are given as:

   
 ∑j mijlij2 − ∑ mijlijS(θij + αi ) ∑ m l C(θ ij ij + αi )
ij  0 0 0
 j j  .  
Mi = − ∑ mijlijS(θij + αi ) ∑j mij 0 ; Ci = − ωi ∑mijlijC(θij + αi ) 0 0 (8)
 j   j 
 ∑ mijlij C(θij + αi ) 0 ∑j mij  − ωi ∑mijlijS(θij + αi ) 0 0
 j   j 

In Eq. (8), C and S are abbreviations for cosine and sine functions, respectively. There are 3k
parameters, mij, θij, lij for j=1, 2,…,k if k point-masses are defined for the ith link. For a
mechanism of n moving links, there will be a total 3kn point-mass parameters. All or some of
these can be taken as the design variables in optimization formulation discussed in the next
section.

ISSN: 2395-1303 http://www.ijetjournal.org Page 54

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

3. FORMULATION OF OPTIMIZATION PROBLEM

Without loosing generalization, the problem for minimizing shaking force and shaking
moment in a planar five-bar mechanism is now formulated on the basis of the dynamics
presented in previous section. For minimizing the inertia forces by redistributing the link
masses, mass and inertia properties of moving links are represented by the dynamically
equivalent systems of point-masses. The point-mass parameters are treated as the design
variables. The five-bar mechanism under consideration is shown in Fig. 3. The links are
numbered as #0, #1, #2, #3 and #4, where link #0 represents the frame to which link #1 and
link #4 are connected. All joints are revolute type. The joints are numbered as 1, 2, 3, 4 and 5
while a0, a1, a2, a3 and a4 represent the link lengths. The fixed inertial frame, OXY, is located
at joint 1, between link #1 and the frame #0.

C2 3 C3
d2 #2 #3
Y θ3
θ2 d3
a2 a3
2
4
#1
C1 θ #4
1 a
1 C4 θ4
a4
d d
1 n01 + n1e n04 + n4e
X 5
f01 f04
-f01
-f04 -n04
-n01 #0
a0
O1 O5
Figure 3. Five-bar mechanism detached from its frame

3.1. Identification of design variables

A system of k equimomental point-masses is used for each link and the corresponding point-
mass parameters are taken as the design variables. Therefore, the 3k-vector of design
variables for the ith link includes point-mass and their locations, and is defined as:

DV1 = [mi1 l i1 θ i1 mi 2 li 2 θi 2 ..... mik l ik θ ik ]T (9)

Hence, the design variable 3kn-vector, DV, for mechanism having n moving links can be
defined as:

DV = DV1T[ DV2T ..... DVnT ]

T
(10)

3.2. Objective function and constraints

For a mechanism in motion, shaking force is the vector sum of the inertia forces, whereas the
shaking moment about any point is the sum of the inertia couples and the moment of the

ISSN: 2395-1303 http://www.ijetjournal.org Page 55

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

inertia forces about that point [12]. In the current problem, the external forces like gravity and
dissipative forces are not considered. Once all the joint reactions are determined, the shaking
force and shaking moment at and about joint 1 are presented as:

f sh = −(f 01 + f 04 ) and nsh = −( n1e + a 0 x f 04 ) (11)

In Eq. (11), f01 and f04 are the reaction forces of the ground on the links 1 and 4, respectively,
while n1e is the driving torque applied at joint 1. a0 is the vector from O1 to O5. Considering
the RMS values of the shaking force, fsh,rms, and the shaking moment, nsh,rms, the optimization
problem is proposed as:

Minimize Z = w1f sh, rms + w2 n sh, rms (12)

Subject to mi, min ≤ ∑ mij ≤ mi,max for i = 1, 2, 3,4 and j = 1, 2,…,k (13)
j

where w1 and w2 are the weighting factors whose values may vary depending on an
application. For example, w1=1.0 and w2=0 if objective is to minimize the shaking force only
and vice-versa. These weighting factors can also be taken as the design variables to get the
most appropraite values for a multi-objective optimization problem. The minimum mass and
inertia, mi,min and Ii,min, of ith link can be defined according to its force bearing capabilities
and link material properties. The solution of this optimization problem finds the values of the
design variables that minimize the objective function Z.

4. SOLUTIONS AND RESULTS

After formulating the balancing problem as an optimization problem, it can be solved by
using either conventional or evolutionary optimization algorithms. The gradient based
conventional algorithms use the gradient information of the objective function with respect to
the design variables. Starting with an initial guess point, these methods converge on the
optimum solution near to the starting point and thus produce local optimum solution.
Genetic algorithm is evolutionary search and optimization algorithm based on the mechanics
of natural genetics and natural selection [24]. This algorithm evaluates only the objective
function and genetic operators - selection, crossover and mutation are used for exploring the
design space. One can specify the initial population, bounds and nonlinear constraints for the
variables in this algorithm. After selection operation, crossover and mutation operators are
used to form the new population. This process is repeated till the convergence criteria is
satisfied [25]. The drawbacks of the GAs are that (1) they require a large amount of
calculation and (2) there is no absolute guarantee that a global solution is obtained. These
drawbacks can be overcome by using parallel computers and by executing the algorithm
several times or allowing it to run longer [26]. The flow chart of the proposed method is
shown in Fig. 4.
The proposed method is applied for the balancing of a five-bar mechanism [17]. In [17], only
shaking force is minimized through conventional optimization method, i.e., non-linear
constraint optimization. The center of mass parameters of moving links were chosen as the
design variables. The natural orthogonal complement method was used for dynamic analysis
of the mechanism. However, the resulting effect on shaking moment and driving torque was
not considered. For the same numerical problem, both shaking force and shaking moment are

ISSN: 2395-1303 http://www.ijetjournal.org Page 56

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

simultaneously minimized in this paper using the global optimization method, i.e., genetic
algorithm.
As shaking force and shaking moment are of different units, these quantities need to be
dimensionless for adding them in a single objective function. For this, the mechanism
parameters are made dimensionless with respect to the parameters of the driving link and
shown in Table 1. For this example, the driving link, i.e. link 1, rotates with a constant speed
of 100 rad/sec.

Design variables:
Point mass parameters and
weighting factors

Objective function and constraints:

Minimize Z = w1f sh,rms + w2 n sh, rms
Subject to mi,min ≤ ∑ mij ≤ mi,max
j

for i = 1, 2, 3, 4 and j = 1, 2,…,k

Solution using “ga” and “gamultiobj” functions in

MATLAB

Optimized values of shaking force

and shaking moment and
corresponding design variables

Optimized mechanism parameters

Figure 4. Flowchart of proposed method

To demonstrate the effectiveness and accuracy of the method, here each link is represented by
three equimomental point-masses. To reduce the dimension of the problem, out of nine
variables, mij, lij, θij, for j=1, 2, 3, for the ith link, five parameters are assigned as: θi1=0;
θi2=2π/3; θi3=4π/3 and li2=li3=li1.
The other four point-mass parameters and weighing factors, namely, mi1, mi2, mi3, li1, w1 and
w2 are brought into the optimization scheme. A MATLAB program was developed using the
equimomental conditions, Eqs. (4)-(7), for finding the dynamically equivalent point-masses
for each link. The resulting point-masses and their locations are shown in Table 2.
Table 1. Dimensionless parameters of standard mechanism
Length Mass Moment Center Center
Link ai mi of of mass of mass
inertia distance location
Iozzi di θi

1 1 1 0.3333 0.5 0
2 5 5 41.6667 2.5 0
3 5 5 41.6667 2.5 0
4 2 2 2.6667 1 0
0 2 - - -

ISSN: 2395-1303 http://www.ijetjournal.org Page 57

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

Table 2. Point-mass parameters

Link mi1 mi2 mi3 li1
1 0.9107 0.0447 0.0447 0.5774
2 4.5534 0.2233 0.2233 2.8868
3 4.5534 0.2233 0.2233 2.8868
4 1.8214 0.0893 0.0893 1.1547

Considering mi,min= 0.75 mi0 , mi,max= 2 mi0 where mi0 is original mass of the ith link, the
optimization problem as explained in Eqs. (12)-(13) is solved using “ga” function in Genetic
Algorithm and Direct Search Toolbox of MATLAB [27]. The original values of point-mass
parameters are taken as the initial population and the algorithm was run for 100 generations.
The comparison of original values with optimum values of the shaking force and shaking
moment obtained using genetic algorithm are presented in Table 3 and Fig. 5. The optimized
link parameters are found by using the equimomental conditions presented in Eqs. (4)-(7) and
shown in Table 4.
Table 3. RMS values of dynamic quantities of standard and optimized mechanisms
RMS values of dimensionless
dynamic quantities
Shaking Shaking
force moment
Standard
2388 21913
value
Genetic 1603 11214
algorithm (-32.87%) (-48.82%)
The values in the parenthesis denote percentage increment/decrement with respect to corresponding RMS values of the
standard mechanism

Table 4. Dimensionless parameters of balanced mechanism

Length Mass Moment Center Center
Link ai mi of of mass of mass
inertia distance location
Iozzi di θi

1 1 1.8783 2.1803 0.6767 356.00

2 5 4.3356 15.6911 1.1160 16.10
3 5 4.6127 9.6128 1.1681 358.12
4 2 3.3909 0.0812 0.0839 331.92
0 2 - - - -

By using the genetic algorithm, the reduction of 32.87% and 48.82% were found in the
values of shaking force and shaking moment, respectively.

ISSN: 2395-1303 http://www.ijetjournal.org Page 58

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

4
9000 x 10
6
Original value Original value
8000 GA value 4 GA value

Normalised Shaking Moment

Normalised Shaking Force 7000 2

6000 0

5000 -2

4000 -4

3000 -6

2000 -8

1000 -10

0 -12
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Time (sec) Time (sec)

(a) Shaking force (b) Shaking moment

Figure 5. Variations of shaking force and shaking moment for complete cycle

Note that the above results are obtained using single objective function mentioned in Eq. (12).
Moreover, this problem can be solved by considering the shaking force and shaking moment
as two objective functions. The multi-objective optimization, also known as vector
optimization, is the procedure used for simultaneous minimization or maximization of more
than one objective function. Various nonlinear multi-objective optimization methods are
surveyed in [28]. The objective function for the posed problem is defined as:

Minimize Z m = [f sh,rms , n sh, rms ]T (14)

This problem is solved using “gamultiobj” function in Genetic Algorithm and Direct Search
Toolbox of MATLAB. This function, Zm, finds the minimum using genetic algorithm and
creates a set of non-dominated solution set known as Pareto front for objectives, i.e., the
shaking force and shaking moment. The values of genetic operators used are:

Selection function = Stochastic uniform

elite count = 2
crossover fraction = 0.8
migration fraction = 0.2 -10
function tolerance = 1 x 10-10
constraint tolerance = 1 x 10
All Pareto solutions are optimum as no other solutions in the entire design space is available
which is better than these solutions when all the objectives are considered. The Pareto front is
shown in Fig. 6 which presents the multiple optimum solutions for the considered problem.
The values of the objective function and corresponding design variables associated with each
point of this curve are also available in the solution. The optimum values of weighting factors
are found as 0.5 each. The optimum values of point-mass parameters are given in Table 5.
Table 5. Optimum point-mass parameters
Link mi1 mi2 mi3 li1
1 1.4107 0.1863 0.2813 1.0774
2 3.0743 1.0380 0.2233 1.9024
3 4.0245 0.2233 0.3649 1.4436

ISSN: 2395-1303 http://www.ijetjournal.org Page 59

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

4 2.2123 0.0893 1.0893 0.1547

4
x 10
1.4
GA - multi-objective
1.35 GA - single objective

Normalised Shaking Moment

1.3

1.25

1.2

1.15

1.1

1.05
1560 1580 1600 1620 1640 1660
Normalised Shaking Force

Figure 6. Pareto front

The results obtained by using single and multi-objective optimization methods are shown in
Fig. 6. For the optimum solutions lying on Pareto front, the decrease in value of one objective
increases value of the other one. Hence, the mechanism designer has several choices to choose
the solution as per the specific requirement.

5. CONCLUSIONS
An optimization method for dynamic balancing of five-bar planar mechanisms is presented
in this paper using the concept of the equimomental system of point-masses for the rigid
body. The dynamic equations of motion are formulated systematically in the parameters
related to the equimomental point-masses. Using these equations, the optimization problem
is formulated for the minimization of the shaking force and shaking moment as single
objective and multi-objective function. With optimum weighting to the objectives, 32.87%
and 48.82% reduction is achieved in shaking force and shaking moment, respectively. The
problem is also formulated as a multi-objective optimization problem for which Pareto front
provides better insight over the combinations of shaking force and shaking moment. The
formulation presented in this paper is simple, easy to implement and it can be applied for
multi-loop planar and spatial mechanisms also.

REFERENCES
[1] Berkof, R.S.; Lowen, G.G.: A New Method for Completely Force Balancing Simple
Mechanisms. ASME Journal of Engineering for Industry, Vol. 91, No. 1, pp. 21-26,
1969.
[2] Tepper, F.R.; Lowen, G.G.: General Theorems Concerning Full Force Balancing of
Planar Mechanisms by Internal Mass Redistribution. ASME Journal of Engineering for
Industry, Vol. 94, No. 3, pp. 789-796, 1972.
[3] Walker, M.J.; Oldham, K.: A General Theory of Force Balancing Using Counterweights.

ISSN: 2395-1303 http://www.ijetjournal.org Page 60

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

Mechanism and Machine Theory, Vol. 13, pp. 175-185, 1978.

[4] Ouyang, P.R.; Li, Q.; Zhang, W.J.: Integrated Design of Robotic Mechanisms for Force
Balancing and Trajectory Tracking. Mechatronics, Vol. 13, pp. 887-905, 2003.
[5] Lowen, G.G.; Tepper, F.R.; Berkof, R.S.: The Quantitative Influence of Complete Force
Balancing on the Forces and Moments of Certain Families of Four-Bar Linkages.
Mechanism and Machine Theory, Vol. 9, pp. 299-323, 1974.
[6] Carson, W.L.; Stephenes, J.M.: Feasible Parameter Design Spaces for Force and Root-
Mean-Square Moment Balancing an In-line 4R 4-Bar Synthesized for Kinematic Criteria.
Mechanism and Machine Theory, Vol. 13, pp. 649-658, 1978.
[7] Hains, R.S.: Minimum RMS Shaking Moment or Driving Torque of a Force-Balanced
Mechanism Using Feasible Counterweights. Mechanism and Machine Theory, Vol. 16,
pp. 185-190, 1981.
[8] Arakelian, V.H.; Smith, M.R.: Complete Shaking Force and Shaking Moment Balancing
of Mechanisms. Mechanism and Machine Theory, Vol. 34, pp. 1141-1153, 1999.
[9] Esat, I.; Bahai, H.: A Theory of Complete Force and Moment Balancing of Planar
Linkage Mechanisms. Mechanism and Machine Theory, Vol. 34, pp. 903-922, 1999.
[10] Feng, G.: Complete Shaking Force and Shaking Moment Balancing of 17 Types of
Eight-bar Linkages Only With Revolute Pairs. Mechanism and Machine Theory, Vol. 26,
No. 2, pp. 197-206, 1991.
[11] Chaudhary, H.; Saha, S.K.: Balancing of Shaking Forces and Shaking Moments for
Planar Mechanisms Using the Equimomental Systems. Mechanism and Machine Theory,
Vol. 43, pp. 310–334, 2008.
[12] Lee, T.W.; Cheng, C.: Optimum Balancing of Combined Shaking Force, Shaking
Moment, and Torque Fluctuations in High Speed Mechanisms. ASME Journal of
Mechanisms, Transmissions, and Automation in Design, Vol. 106, No. 2, pp. 242-251,
1984.
[13] Chaudhary, K.; Chaudhary, H.: Concept of Equimomental System for Dynamic
Balancing of Mechanisms. International Conference on Automation and Mechanical
Systems, pp. 124-132, Lingaya’s University, Faridabad, India 2013.
[14] Mariappan, J.; Krishnamurty, S.: A Generalised Exact Gradient Method for Mechanism
Synthesis. Mechanism and Machine Theory, Vol. 31, No. 4, pp. 413-421, 1996.
[15] Nokleby, S.B.; Podhorodeski, R.P.: Optimization-based Synthesis of Grashof Geared
Five-bar Mechanism. ASME Journal of Mechanical Design, Vol. 123, pp. 529-534,
2001.
[16] Li, Z.: Sensitivity and Robustness of Mechanism Balancing. Mechanism and Machine
Theory, Vol. 33, No. 7, pp. 1045-1054, 1998.
[17] Ilia, D.; Sinatra, R.: A Novel Formulation of the Dynamic Balancing of Five-Bar
Linkages with Application to Link Optimization. Multibody Systems Dynamics, Vol. 21,
pp. 193-211, 2009.
[18] Alici, G.; Shirinzadeh, B.: Optimum Dynamic Balancing of Planar Parallel Manipulators
Based on Sensitivity Analysis. Mechanism and Machine Theory, Vol. 41, pp. 1520-1535,
2006.
[19] Farmani, M.R.; Jaamialahmadi, A.; Babaie, M.: Multiobjective Optimization for Force

ISSN: 2395-1303 http://www.ijetjournal.org Page 61

 International Journal of Engineering and Techniques - Volume 3 Issue 6, Nov - Dec 2017

and Moment Balance of a Four-bar Mechanism Using Evolutionary Algorithms. Journal

of Mechanical Science and Technology, Vol. 25, No. 12, pp. 2971-2977, 2011.
[20] Erkaya, S.: Investigation of Balancing Problem for A Planar Mechanism Using Genetic
Algorithm. Journal of Mechanical Science and Technology, Vol. 27, No. 7, pp. 2153-
2160, 2013.
[21] Guo, G.; Morita, N.; Torii, T.: Optimum Dynamic Design of Planar Linkage Using
Genetic Algorithms. JSME International Journal Series C, Vol. 43, No. 2, pp. 372-377,
2000.
[22] Routh, E.J.: Treatise on the Dynamics of a System of Rigid Bodies, Elementary Part I.
New York, USA: Dover Publication Inc., 1905.
[23] Chaudhary, H.; Saha, S.K.: Balancing of Four-bar Mechanisms Using Maximum
Recursive Dynamic Algorithm. Mechanism and Machine Theory, Vol. 42, No. 2, pp.
216-232, 2007.
[24] Deb, K.: Optimization for Engineering Design – Algorithms and examples. New Delhi:
PHI Learning Private Limited, 2010.
[25] Gao, Y.; Shi, L.; Yao, P.: Study on Multi-objective Genetic Algorithm. 3rd World
Congress on Intelligent Control and Automation, Hefei, P R China 2000.
[26] Arora, J.S.: Introduction to optimum design. Singapore: McGraw-Hill Book Company,
1989.
[27] MATLAB Optimization Toolbox, version 7.7.0.471 (R2008b).
[28] Marler, R.T.; Arora, J.S.: Survey of Multi-objective Optimization Methods for
Engineering. Structural and Multidisciplinary Optimization, Vol. 26, No. 6, pp. 369-395,
2004.

ISSN: 2395-1303 http://www.ijetjournal.org Page 62