Energy Methods: Lagrange’s Equations

Lecture L20 - Energy Methods: Lagrange’s Equations The motion of particles and rigid bodies is governed by Newton’s law. In this section, we will derive an alternate approach, placing Newton’s law into a form particularly convenient for multiple degree of freedom systems or systems in complex coordinate systems. This approach results in a set of equations called Lagrange’s equations. They are the beginning of a complex, more mathematical approach to mechanics called analytical dynamics. In this course we will only deal with this method at an elementary level. Even at this simplified level, it is clear that considerable simplification occurs in deriving the equations of motion for complex systems. These two approaches–Newton’s Law and Lagrange’s Equations–are totally compatible. No new physical laws result for one approach vs. the other. Many have argued that Lagrange’s Equations, based upon conservation of energy, are a more fundamental statement of the laws governing the motion of particles and rigid bodies. We shall not enter into this debate. Derivation of Lagrange’s Equations in Cartesian Coordinates We begin by considering the conservation equations for a large number (N) of particles in a conservative force field using cartesian coordinates of position xi . For this system, we write the total kinetic energy as T = M � 1 mi ẋ2i . 2 n=1 (1) where M is the number of degrees of freedom of the system. For particles traveling only in one direction, only one xi is required to define the position of each particle, so that the number of degrees of freedom M = N . For particles traveling in three dimensions, each particle requires 3 xi coordinates, so that M = 3 ∗ N . The momentum of a given particle in a given direction can be obtained by differentiating this expression with respect to the appropriate xi coordinate. This gives the momentum pi for this particular particle in this coordinate direction. ∂T = pi ∂ẋi (2) The time derivative of the momentum is d dt � ∂T ∂ẋi � = mi ẍi 1 (3) For a conservative force field, the force on a particle is given by the derivative of the potential at the particle position in the desired direction. Fi = − ∂V ∂xi (4) From Newton’s law we have Fi = dpi dt (5) Equating like terms from our manipulations on kinetic energy and the potential of a conservative force field, we write d dt � ∂T ∂ẋi � =− ∂V ∂xi (6) Now we make use of the fact that ∂T =0 ∂xi (7) ∂V =0 ∂ẋi (8) and Using these results, we can rewrite Equation (6) as � � d ∂(T − V ) ∂(T − V ) − =0 dt ∂x˙ i ∂xi (9) We now define L = T − V : L is called the Lagrangian. Equation (9) takes the final form: Lagrange’s equations in cartesian coordinates. d dt � ∂L ∂ẋi � − ∂L =0 ∂xi (10) where i is taken over all of the degrees of freedom of the system. Before moving on to more general coordinate systems, we will look at the application of Equation(10) to some simple systems. Mass-spring System We first consider a simple mass spring system. This is a one degree of freedom system, with one xi . Its kinetic energy is T = 1/2mẋ2 ; its potential is V = 1/2kx2 ; its Lagrangian is L = 1/2mẋ2 −1/2kx2 . Applying Equation (10) to the Lagrangian of this simple system, we obtain the familiar differential equation for the mass-spring oscillator. m d2 x + kx = 0 dt2 (11) Clearly, we would not go through a process of such complexity to derive this simple equation. However, let’s consider a more complex system, governed by the same laws. 2 This is a 2 degree of freedom system, governed by 2 differential equations. The number of springs for this configuration is 3. These governing equations could be obtained by applying Newton’s Law to the force balance that exists at each mass due to the deflection of the springs as was done in Lecture 19. The deflection of springs 1 and 3 are influenced by the boundary condition at either end of the slot; in this case the deflection is zero. The governing equations can also be obtained by direct application of Lagrange’s Equation. This approach is quite straightforward. The expression for kinetic energy is T = 2 � 1 mi ẋ2i ; 2 n=1 (12) the expression for the potential is V = 1/2k1 x21 + 1/2k2 (x2 − x1 )2 + 1/2k3 x22 Applying Lagrange’s equation to T − V = L � � d ∂L − dt ∂ ẋ1 � � d ∂L − dt ∂ ẋ2 ∂L =0 ∂x1 ∂L =0 ∂x2 (13) (14) (15) we obtain the governing equations as d2 x1 dt2 d2 x2 m2 2 dt m1 = −k1 x1 + k2 (x2 − x1 ) (16) = −k2 (x2 − x1 ) − k3 x2 (17) Clearly, for multi degree of freedom systems, this approach has advantages over the force balancing approach using Newton’s law. Extension to General Coordinate Systems A significant advantage of the Lagrangian approach to developing equations of motion for complex systems comes as we leave the cartesian xi coordinate system and move into a general coordinate system. An 3 example would be polar coordinates where for a two-dimensional position of a mass particle, x1 and x2 could be given by r and θ. A two-degree of freedom system remains two-degree so that the number of coordinate variables required remains two. r and θ and their counterparts in other coordinate systems will be referred to as generalized coordinates. We introduce quite general notation for the relationship between the n cartesian variables of position xi and their description in generalized coordinates. (For some systems, the number of generalized coordinates is larger than the number of degrees of freedom and this is accounted for by introducing constraints on the system. This is an important part of the discussion of the Lagrange formulation. We shall not however develop these relations but will work directly with the number of variables equal to the number of degrees of freedom of the system.) We express the cartesian variable xi using generalized coordinates qj . (Polar coordinates r, θ would be an example.) xi = xi (q1 , ..qj , ...qn ). (18) In the general case, each xi could be dependent upon every qj . What is remarkable about the Lagrange formulation, is that (10) holds in a general coordinate system with xi replaced by qi . d dt � ∂L ∂q̇i � − ∂L =0 ∂qi (19) Before showing how this result can be derived from Newton’s Law, we show two applications in polar coordinates to demonstrate the power of the approach. Simple Pendulum by Lagrange’s Equations We first apply Lagrange’s equation to derive the equations of motion of a simple pendulum in polar coor­ dinates. This is a one degree of freedom system. However, it is convenient for later analysis of the double pendulum, to begin by describing the position of the mass point m1 with cartesian coordinates x1 and y1 and then express the Lagrangian in the polar angle θ1 . Referring to a) in the figure below we have x1 = h1 sin θ1 (20) y1 = −h1 cos θ1 (21) 1 1 m1 (ẋ21 + ẏ12 ) = m1 h21 θ̇12 2 2 (22) so that the kinetic energy is T = The potential energy is V = m1 gy1 = −m1 gh1 cos θ (23) The Lagrangian is L=T −V = 1 m1 h21 θ̇12 + m1 gh1 cos θ1 2 4 (24) Applying (10) with q1 = θ1 , we obtain the differential equation governing the motion. m1 h21 θ̈1 + m1 gh1 sin θ1 = 0 (25) Again, for such a simple system, we would typically not go through this formalism to obtain this result. However, this framework will enable us to derive the equations of motion for the more complex systems such as the double pendulum shown in b). Double Pendulum by Lagrange’s Equations Consider the double pendulum shown in b) consisting of two rods of length h1 and h2 with mass points m1 and m2 hung from a pivot. This systems has two degrees of freedom: θ1 and θ2 . To apply Lagrange’s equations, we determine expressions for the kinetic energy and the potential as the system moves in angular displacement through the independent angles θ1 and θ2 . From the geometry we have The kinetic energy is T = 1 2 m1 � x1 = h1 sin θ1 (26) y1 = −h1 cos θ1 (27) x2 = h1 sin θ1 + h2 sin θ2 (28) y2 = −h1 cos θ1 − h2 cos θ2 (29) � � � ẋ21 + ẏ12 + 12 m2 ẋ22 + ẏ22 . Expressed in variables θ1 and θ2 , the kinetic energy of the system is T = � � 1 1 m1 h21 θ̇12 + m2 h21 θ̇12 + h22 θ̇22 + 2h1 h2 θ̇1 θ̇2 cos(θ1 − θ2 ) 2 2 5 (30) The potential energy of the system is V = m1 gy1 + m2 gy2 = −(m1 + m2 )gh1 cos θ1 − m2 gh2 cos θ2 . (31) The Lagrangian is then L = T −V = 1 1 (m1 +m2 )h21 θ̇12 + m2 h22 θ̇22 +m2 h1 h2 θ̇1 θ̇2 cos(θ1 −θ2 )+(m1 +m2 )gh1 cos θ1 +m2 gh2 cos θ2 (32) 2 2 Since the generalized coordinates are now θ1 and θ2 , Lagrange’s equation becomes � � d ∂L ∂L − =0 dt ∂θ̇i ∂θi (33) for both i = 1 and i = 2. � � d ∂L − dt ∂θ̇1 � � d ∂L − dt ∂θ̇2 ∂L =0 ∂θ1 ∂L =0 ∂θ2 Working out the details, we have � � d ∂L = (m1 + m2 )h21 θ̈1 + m2 h1 h2 θ̈2 cos(θ1 − θ2 ) − m2 h1 h2 θ̇2 sin(θ1 − θ2 )(θ̇1 − θ̇2 ) dt ∂θ̇1 ∂L = −h1 g(m1 + m2 ) sin(θ1 ) − m2 h1 h2 θ̇1 θ̇2 sin(θ1 − θ2 ) ∂θ1 � � d ∂L = m2 h22 θ¨2 + m2 h1 h2 θ¨1 cos(θ1 − θ2 ) − m2 h1 h2 θ̇1 sin(θ1 − θ2 )(θ̇1 − θ̇2 ) dt ∂θ̇2 ∂L = −h2 gm2 sin(θ2 ) + m2 h1 h2 θ̇1 θ̇2 sin(θ1 − θ2 ) ∂θ2 (34) (35) (36) (37) (38) (39) From Equation(35-39), we obtain the final form of the governing equations for the double pendulum (m1 + m2 )h1 θ̈1 + m2 h2 θ̈2 cos(θ1 − θ2 ) + m2 h2 θ̇22 sin(θ1 − θ2 ) + g(m1 + m2 ) sin θ1 = 0 (40) m2 h2 θ̈2 + m2 h1 θ̈1 cos(θ1 − θ2 ) − m2 h1 θ̇12 sin(θ1 − θ2 ) + m2 g sin θ2 = 0 (41) The double pendulum is a system of great interest, displaying conventional linear multi degree of freedom sys­ tem behavior for small θ1 and θ2 , but displaying chaotic behavior for large θ. A chaotic system is a determin­ istic system that exhibits great sensitivity to the initial conditions: the ”butterfly” effect. A simulation of the motion of a double pendulum is available on http://scienceworld.wolfram.com/physics/DoublePendulum.html. For a particular choice of initial conditions, the position of m2 with time is shown in the figure. 6 Derivation of Lagrange’s Equation for General Coordinate Systems We now follow the earlier procedure we used to derive Lagrange’s equation from Newton’s law but using generalized coordinates instead of cartesian coordinates. (See additional reading: Slater and Frank and/or Marion and Thorton.) The cartesian variable are expressed in terms of the generalized coordinates as xi = xi (q1 , ..qj , ...qn ). (42) To obtain the velocity ẋi , we take the time derivative of (42) applying the chain rule. ẋi = n � ∂xi j=1 ∂qj q̇j . (43) Taking the partial derivative of (43) with respect to q̇j we obtain a relation between these two derivatives, ∂ẋi ∂xi = , ∂q̇j ∂qj (44) a result which we shall need shortly. In deriving equation (44), we take advantage of the fact that since the generalized coordinates qi are independent, ∂qi ∂qj = 0 for i = � j. Following the approach leading to Equations (2-10), we define a generalized momentum as pi = ∂T ∂q̇i (45) M � 1 mj ẋ2j as given by Equation (1). 2 n=1 We are now ready to express Newton’s law in the generalized coordinates qi that we have introduced. with T = 7 For the generalized momentum we have n pi = n � � ∂T ∂ẋj ∂xj = mj ẋj = mj ẋj , ∂q̇i ∂q̇ ∂qi i j=1 j=1 (46) where we have made use of (44). For a conservative force field, the work done during a displacement dxi is given by dW = N � Fi dxi (47) i=1 or expressed in the generalized coordinates qj dW = N � Fi dxi = N � N � i=1 Fi j=1 i=1 ∂xi dqj ∂qj (48) We identify N � Fi i=1 ∂xi dqj ∂qj (49) as the work done by a ”displacement” through dqj and define a generalized force Qj as Qj = N � ∂xi ∂qj (50) Qj dqj (51) Fi i=1 so that the work is expressed as. dW = N � j=1 For a conservative system, the work done by a small displacement dqj is dW = −dV = − ∂V dqj , ∂qj (52) where V is the potential function expressed in the coordinate system of the generalized coordinates and ∂V ∂qj is the change in potential due to a change in the generalized coordinate qj . So that the generalized force is Qj = − ∂V ∂qj (53) in analogy with (4). We now examine the time derivative of the generalized momentum, pi . N � dpi d ∂T ∂xj d ∂xj = ( )= (mj ẍj + mj ẋj ). dt dt ∂qi ∂q dt ∂qi i j=1 (54) Since ∂xj /∂qi is a function of the q � s which are functions of time, we have by the chain rule N � ∂ 2 xj d ∂xj ( )= q̇k dt ∂qi ∂qi ∂qk (55) k=1 From Newton’s law, mj ẍj = Fj , so that the first term in (54) is given by N � j=1 mj ẍj ∂xj = Qi ∂qi 8 (56) Now let us consider the second term of (54). Consider the expression for ∂T /∂qi and use (43, 44). N ∂T ∂ẋj ∂ � ∂xi = mj ẋj = mj ẋj q̇k ∂qi ∂qi ∂qi ∂qk (57) k=1 This is precisely the second term in equation (54). We have now identified simpler forms for the two expressions in the equation for the time derivative of the generalized momentum. We may now write dpi d = dt dt � ∂T ∂q̇i � = Qi + ∂T ∂V ∂T =− + ∂qi ∂qi ∂qi (58) Since for a conservative system, the potential is independent of the velocities, we can place this equation into final form by defining the Lagrangian as L = T − V to obtain the final form of Lagrange’s equation as d dt � ∂L ∂q̇i � − ∂L =0 ∂qi (59) in agreement with (19). Equation (59) is Lagrange’s Equations in generalized coordinates. In our previous example, we applied this equation to simple and double pendulums in polar coordinates using q1 = θ1 and q2 = θ2 . What is significant about this equation, adding to its power, is that each i equation contains only derivatives with respect to that qi and q̇i . ADDITIONAL READING Slater and Frank, Mechanics, Chapter IV Marion and Thorton, Chapter 7 9