Scribd
Upload
21 views21 pages

Lecture 11B

The document provides an overview of the maintenance and implementation of PID control systems in industrial settings, including evaluations and examples. It also discusses Monte Carlo methods and Bayesian filtering for non-linear systems with uncertainties, focusing on particle filtering techniques. Key components of Bayesian approaches are outlined, emphasizing state and measurement vectors, along with the importance of re-sampling in sequential importance particle filters.

Uploaded by

Savitri Shrivastava
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
21 views21 pages

Lecture 11B

The document provides an overview of the maintenance and implementation of PID control systems in industrial settings, including evaluations and examples. It also discusses Monte Carlo methods and Bayesian filtering for non-linear systems with uncertainties, focusing on particle filtering techniques. Key components of Bayesian approaches are outlined, emphasizing state and measurement vectors, along with the importance of re-sampling in sequential importance particle filters.

Uploaded by

Savitri Shrivastava
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 21

Recap

An Overview of Maintenance of
PID Control Systems in
Industry
Maintenance of PID Control Systems in Industry
Understanding the PID Controls
Evaluating Controllers
A Few Industrial Examples
A Few Industrial Examples
A Few Industrial Examples
Implementation in Digital Domain
Introduction to Non-Linear Estimation
Lecture-11-B

randper
m
Monte Carlo Methods

Less Theory, More Simulation


E x  

x f x  dx

1 N
E x  
N
 xi
i 1
Non-Linear Filter with Non-Gaussian Uncertainties in the Bayesian Framework

Update
Propagation of a Conditional Density Function

(k-1)+ (k)- (k)+

k-1 Prediction
k

z(k-1) z(k)
Bayesian Filtering/Estimation for Non-Linear Systems with
Non-Gaussian Process & Measurement Uncertainties

State Variable Representation of Dynamical Systems, which are Strongly Non-Linear


Components in a Bayesian Approach

• State Vector : xk

• Measurement Vector: zk

• Initial pdf: p(x0)

• Likelihood pdf: (from model) p(zk|xk)

• Prior pdf: p(xk|z1:k-1)

• Posterior pdf: p(xk|z1:k )


Particle Filtering --- Similar Recursions, but with a Difference

Instantiating
Re-sampling
X(k) = f[x(k-1), Bu(k-1)] + w(k-1) Update

(k-1)+ (k)- (k)+

k-1 k

z(k-1) X(k) = Ax(k-1) + Bu(k-1) z(k)


Propagating from (k-1) to (k) to (k+1)
Re-sampling
Sequential Importance Re-sampling/Particle Filter

You might also like