Energy function of 2D and 3D dynamical systems
Jean-Marc Ginoux
a,∗, Riccardo Meucci
a,
Jaume Llibre
b, Julien Clinton Sprott
c
aNational Institute of Optics - CNR,
Florence, Italy
bDepartament de Matemàtiques,
Universitat Auṭnoma de Barcelona, 08193 Bellaterra,
Barcelona, Spain
cUniversity of Wisconsin, 1150
University Avenue Madison, WI 53706-1390, USA
It is far well-known that energy function of a
two-dimensional autonomous dynamical system can be simply
obtained by multiplying its corresponding second-order
ordinary differential equation, i.e., its equation of
motion by the first time derivative of its state variable. In
the nineties, one of us (J.C.S.) stated that a
threedimensional autonomous dynamical system can be also
transformed into a third-order ordinary differential
equation of motion todays known as jerk equation. Although a
method has been developed during these last
decades to provide the energy function of such
three-dimensional autonomous dynamical systems, the question
arose to determine by which type of term, i.e., by the first
or second time derivative of their state variable, the
corresponding jerk equation of these systems should be
multiplied to deduce their energy function. We prove
in this work that the jerk equation of such systems must be
multiplied by the second time derivative of the
state variable and not by the first like in dimension two. We
then provide an interpretation of the new term
appearing in the energy function and called jerk energy. We
thus established that it is possible to obtain the
energy function of a three-dimensional dynamical system
directly from its corresponding jerk equation. Two and
three-dimensional Van der Pol models are then used to
exemplify these main results. Applications to Lorenz
and Chua’s models confirms their validity.