Article Preview
Top2. The Action Functional In Different Sets Of Admissible Functions
Let us consider the action functional for a single particle moving along x-axis (1) where L(t,x,) is the Lagrangian. If x(t) is the real physical curve described by the particle (trajectory), such that x(t1)=a, x(t2)=b, we consider the functional (1) in the following sets of admissible functions
(2)Here D1(t1,t2) is the space of all functions defined on the interval [t1,t2] which are continuous and have continuous first derivatives; h(t) is an arbitrary variable function, h0(t) and h1(t) are arbitrary but fixed functions; α is a real variable parameter. In Figure 1 is reported an example of x2-trajectory.
Figure 1. Example of the trajectory x2(t,α)
The increment of action functional corresponding to the increment h(t) of x(t) is written as (3) where δS[x;h] is the differential of the functional and ε→0 for ||h||→0. The last action principle states that in the set Γ1 we have
(4)We now observe that (δS(1)[x]=0)→ (δS(2)[x]=0)→ (δS(3)[x]=0).
In fact an explicit form of the formula (4) can be derived by using the Taylor’s theorem. So we have (5) where the subscript denotes partial derivatives with respect to the corresponding arguments, and the dots denote terms of order higher than 1 relative to h and .
In this manner the formula (4) can be written as
(6)By considering the action functional in the set Γ2 we have obviously
(7)So we have
(8)We can also prove the implication (9) by showing that (Gelfand et al., 1963)