Systems

This section of the manual describes the systems types that are used in this library.

MathematicalSystems.jl provides some convenience types and methods to work with mathematical systems models. Every system inherits from AbstractSystem.

Linear systems

Two commonly used types of systems are discrete and continuous systems.

Discrete systems. A discrete system consists of a matrix representing the system dynamics, a set of initial states, a set of nondeterministic inputs, and a discretization step δ.

Continuous systems. A continuous system consists of a matrix representing the system dynamics, a set of initial states, and a set of nondeterministic inputs.

Nondeterministic inputs

The above systems may contain nondeterministic inputs, which are wrapped in special types. Every nondeterministic input representation inherits from NonDeterministicInput.

The inputs are closely related to a DiscreteSystem in the sense that for each discrete time step the input set may change. We support iteration through the inputs over time.

Constant nondeterministic inputs

Constant nondeterministic inputs are chosen from a set of values that does not change over time. Note that, while the set is constant, the inputs themselves vary over time.

Time-varying nondeterministic inputs

Time-varying nondeterministic inputs are chosen from a set of values that changes over time (with each time step).

Second order systems

A second order system is one of the form

$$$Mx''(t) + Cx'(t) + Kx(t) = f(t)$$$

where $x(t) ∈ \mathbb{R}^n$ is the state vector and $f : \mathbb{R} \to \mathbb{R}^n$ is the forcing term. Here $M$, $C$ and $K$ are often called the mass matrix, viscosity matrix and stiffness matrix respectively. These names are adopted from physical applications, particularly from structural mechanics. Assuming that the matrix $M$ is invertible, we can transform the second order system to a first order system introducing auxiliary variables, $x̃(t) = [x(t),~v(t)]^T$, where $v(t) := x'(t)$ is the vector of velocities. Then,

$$$x̃(t)' = Ax̃(t) + Bf(t)$$$

where

$$$A = \begin{pmatrix} 0 && I \\ -M^{-1}K && -M^{-1}C \end{pmatrix},\qquad B = \begin{pmatrix} 0 \\ M^{-1} \end{pmatrix}$$$

See the SecondOrder documentation in MathematicalSystems.jl for additional details in second order ODEs types.

Note

A similar relation can be obtained using the alternative convention $[v(t),~x(t)]^T$. Use derivatives_first=true in the normalize function to swap between these conventions (it is set to false by default).