In the study of dynamic systems, the discussion of stability often becomes the key. Whether it is differential or difference equations, different types of stability are crucial to our understanding of the behavior of the system. The most important one is the stability of the solution near the equilibrium point. All this is thanks to the Russian mathematician Alexander Lyapunov, whose Lyapunov stability theory played a foundational role in this regard.
If the solution of the system continues to approach a certain equilibrium point within a certain range of confidence, then the equilibrium point is called Lyapunov stable.
Simply put, if the system starts near an equilibrium point and can always remain near it, then this equilibrium point is stable; and if all solutions not only remain near it, but also tend to move towards this equilibrium point , this stability is strengthened into asymptotic stability. Stronger concepts, such as exponential stability, further emphasize the rate of convergence of solutions, providing us with deeper insights into dynamic systems.
Lyapunov's theory can be traced back to his 1892 paper "General Problems of Stability of Motion" at Kharkov University. Sadly, despite the far-reaching impact of his theories, Lyapunov was not widely recognized and respected during his lifetime. Compared with his contributions, the application of this theory in the field of science and technology has actually received belated attention.
His work lay dormant for many years until Nikolai Chetaev rekindled interest in the theory in the 1930s.
After realizing the potential of Lyapunov's stability theory, Chetaev further generalized this idea so that it could be applied to a wider range of nonlinear dynamic systems. Subsequently, with the resurgence of research during the Cold War, the Lyapunov method gained new recognition, especially in guidance systems in the aerospace field, due to its ability to effectively deal with nonlinear problems.
In a continuous time system, when we consider an automatic nonlinear dynamic system, if its equilibrium point
If there is some distance less than
δ
such that the solution remains withinε
as time progresses, then the equilibrium point is stable.
Under appropriate circumstances, stability theory can also be transferred to higher-dimensional manifolds, in what is called structural stability, focusing on the behavior of different but similar solutions. Furthermore, input-to-state stability (ISS) applies Lyapunov's theory to systems with inputs.
In Lyapunov's original work, he proposed two methods to prove stability. The first method involves expanding the solution to prove its convergence, while the second method, which is now called the "direct method", involves measuring the stability of the system by introducing the Lyapunov function. This function is similar to the potential function in classical dynamics and can provide an intuitive explanation of the energy loss of a system from an unstable state to a stable state. If we can find a suitable Lyapunov function, we can prove the stability of the system without relying on the specific physical energy.
As the research on Lyapunov's theory deepens, we begin to face a new problem: How can we better solve the stability problem of dynamic systems in complex environments? Lyapunov's stability theory not only changed our understanding of dynamic systems, but also provided new perspectives and challenges for future research. Does this mean we need to re-examine our definition and application of stability?