AbstractModel An abstract type that should be subtyped for model extensions (linear or non linear)

ContinuousLinearModel

Model linear implementation with AutomationLabs.

** Fields **

• A: state matrix.
• B: input matrix.

ContinuousNonLinearModel

Model non linear implementation with AutoamtionLabs.

** Fields **

• f: the non linear model.
• nbr_state: the state number.
• nbr_input: the input number

DenseNet

An densely connected network architecture type for dynamical system identification problem [ref].

DiscreteLinearModel

Model linear implementation with AutomationLabs.

** Fields **

• A: state matrix.
• B: input matrix.

DiscreteNonLinearModel

Model non linear implementation with AutoamtionLabs.

** Fields **

• f: the non linear model.
• nbr_state: the state number.
• nbr_input: the input number

Fnn

An feedforward neural network architecture type for dynamical system identification problem [ref].

gru

A gated recurrent unit recurrent neural network architecture type for dynamical system identification problem [ref].

Icnn

An input convex neural network architecture type for dynamical system identification problem [ref].

linear

An linear (Wv –> Ax + Bu) architecture type for dynamical system identification problem [ref].

lstm

A long short-term memory recurrent neural network architecture type for dynamical system identification problem [ref].

NeuralODE An neural neural network ODE architecture type for dynamical system identification problem [ref].

PolyNet

An poly-inception network architecture type for dynamical system identification problem [ref].

Rbf

An radial basis neural network architecture type for dynamical system identification problem [ref].

ResNet

An residual layer network architecture type for dynamical system identification problem [ref].

Rknn1

A runge-kutta neural neural network 1 architecture type for dynamical system identification problem [ref].

Rknn2

A runge-kutta neural neural network 2 architecture type for dynamical system identification problem [ref].

Rknn4

A runge-kutta neural neural network 4 architecture type for dynamical system identification problem [ref].

Rnn

A recurrent neural network architecture type for dynamical system identification problem [ref].

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from non linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from non linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from non linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from non linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from non linear model.

controllersystem_design A function for design the system (model and constraints) with MathematicalSystems from non linear model.

proceed_system

A function for system creation from non-linear discrete or continuous model.

The following variables are mendatories:

• f: a non-linear function.
• nbr_state: the state number.
• nbr_input: the input number.
• variation: continuous or discrete variation.

It is possible to define optional variables kws.

proceed_system

A function for system creation from linear discrete or continuous model.

The following variables are mendatories:

• A: a state matrix.
• B: a input matrix.
• nbr_state: the state number.
• nbr_input: the input number.
• variation: continuous or discrete variation.

It is possible to define optional variables kws.

proceed_system_constraints_evaluation

Function that return the constraints of a AutomationLabsSystem.

** Required fields **

• system: the mathematital system that as in it the julia linear or non-linear function f.

proceed_system_discretization

Function that linearises a system from MathematicalSystems at state and input references. The function uses ForwardDiff package and the jacobian function.

** Required fields **

• system: the continuous mathematital system that as in it the julia linear or non-linear function f.
• sample_time: the sample time for discretization.

proceed_system_evaluation

Function that return the MathematicalSystems type of a AutomationLabsSystem.

** Required fields **

• system: the mathematital system that as in it the julia linear or non-linear function f.

proceed_system_linearization

Function that linearises a system from MathematicalSystems at state and input references. The function uses ForwardDiff package and the jacobian function.

** Required fields **

• system: the mathematital system that as in it the julia non-linear function f.
• state: references state point.
• input: references input point.

proceed_system_model_evaluation

Function that return the types of the model inside the systems from AutomationLabsIdentification.

** Required fields **

• system: the mathematical system that as in it the julia non-linear function f from AutomationLabsIdentification.