9 - Neural Modelling and Control
9 - Neural Modelling and Control
9 - Neural Modelling and Control
and Control
part two
Artificial Intelligence for
Control and Identification
Dr. Wilbert G. Aguilar
Ph.D. in Automatic Control, Robotics and Computer Vision
1. Basic Requirements
2. Classical Modelling
Neural Modelling and Control
3. Neural Modelling
4. Neural Control
1
3. Neural Modelling
Strategies
• Series – Parallel
2
3. Neural Modelling
+ Unknown ym(t)
um(t) +
Plant
Neural Modelling and Control
+ Unknown ym(t)
um(t) +
Plant
Neural Modelling and Control
Fault diagnosis.
5
3. Neural Modelling
Main Problem
6
3. Neural Modelling
+ Unknown ym(t)
um(t) +
Plant
Neural Modelling and Control
7
3. Neural Modelling
Warning
8
3. Neural Modelling
Main Problem
9
3. Neural Modelling
2
⎟
⎜ ⎝ ∂u(t) ⎠ ⎟
∂u(t) ⎝ ⎠
For a nonlinear plant, minimizing MSEu does not imply necessarily
minimizing MSEy
However, when output errors are not correlated with the Jacobian of
the plant, then
⎛ ⎛ ∂y(t) ⎞2 ⎞
Jy ≈ E ⎜ ⎜ ⎟⋅J
⎜ ⎝ ∂u(t) ⎟⎠ ⎟ u
⎝ ⎠
and the Jacobian effect can be incorporated on the learning rate, η
10
3. Neural Modelling
11
3. Neural Modelling
Training phases
i. A direct plant model is built covering all the operations range
Neural Modelling and Control
12
3. Neural Modelling
Advantage
Drawback
13
3. Neural Modelling
Backpropagation-through-time Modelling
z-1
εu(k)
BP
εy(k)
+ –
An emulation of εu(k) is obtained,
backpropagating εy(k) towards the input layer of the ANN plant model
14
3. Neural Modelling
Backpropagation-through-time Modelling
Remarks
This method assumes that input u(t) has a linear effect on the
Neural Modelling and Control
15
3. Neural Modelling
BP