Scribd
Upload
387 views16 pages

7 - 1 - Empirical Model PDF

This document describes procedures for building empirical models of dynamic systems using experimental data. It discusses using statistical methods like linear regression or graphical methods to estimate the parameters of simple models like first-order plus time delay models. Graphical methods involve analyzing the process reaction curve from a step change experiment to determine the steady state gain and time constant. The document provides an example of applying these methods to estimate a first-order model for temperature response in a continuous stirred-tank reactor. It also discusses some drawbacks of simple first-order models.

Uploaded by

Ananda Cahya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
387 views16 pages

7 - 1 - Empirical Model PDF

This document describes procedures for building empirical models of dynamic systems using experimental data. It discusses using statistical methods like linear regression or graphical methods to estimate the parameters of simple models like first-order plus time delay models. Graphical methods involve analyzing the process reaction curve from a step change experiment to determine the steady state gain and time constant. The document provides an example of applying these methods to estimate a first-order model for temperature response in a continuous stirred-tank reactor. It also discusses some drawbacks of simple first-order models.

Uploaded by

Ananda Cahya
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

EMPIRICAL MODEL

EMPIRICAL MODEL BUILDING PROCEDURE

A priori knowledge Start

Experimental Design
Not just
process
control Plant Experimentation

Determine Model Structure

Parameter Estimation

Diagnostic Evaluation
Alternative
data Model Verification

Complete
Method to estimate parameter
in Empirical model

■ Statistic (Linear regression or nonlinear


regression)
■ Graphical Method (Process Reaction Curve)
– First method (slope method)
– Second method  use specific point

These method can be used to build empirical


model for:
• First Order system
• FOPTD
• Second order system or SOPTD
Simple Linear Regression
• As an illustrative example, consider a simple linear model
between an output variable y and input variable u,
y  β1  β 2u  ε
where β1 and β2 are the unknown model parameters to be
Chapter 7

estimated and e is a random error.

• Predictions of y can be made from the regression model,


𝑦 = β1 + β2 𝑢 (7−3)
where 𝛽1 and 𝛽2 denote the estimated values of b1 and b2,
and 𝑦 denotes the predicted value of y.
• Let Y denote the measured value of y. Each pair of (ui, Yi)
observations satisfies:
𝑌𝑖 = β1 + β2 𝑢𝑖 + ε𝑖 (7−1)
The Least Squares Approach
• The least squares method is widely used to calculate the
values of b1 and b2 that minimize the sum of the squares of
the errors (SSE) for an arbitrary number of data points, N:
𝑁

𝑆𝑆𝐸 = 𝑒𝑖2
𝑖=1
where the 𝑖−th residual, 𝑒𝑖 , is defined as,
𝑒𝑖 = 𝑌𝑖 − 𝑦𝑖

𝑁 𝑁
2
𝑆𝑆𝐸 = e2𝑖 = 𝑌𝑖 − β1 − β2 𝑢𝑖 (7−2)
𝑖=1 𝑖=1
The Least Squares Approach (continued)
• The least squares solution that minimizes the sum of
squared errors, SSE, is given by:
𝑆𝑢𝑢 𝑆𝑦 − 𝑆𝑢𝑦 𝑆𝑢
β1 = (7−5)
𝑁𝑆𝑢𝑢 − 𝑆𝑢 2

𝑁𝑆𝑢𝑦 − 𝑆𝑢 𝑆𝑦
β2 = (7−6)
𝑁𝑆𝑢𝑢 − 𝑆𝑢 2

where:
𝑁 𝑁 𝑁 𝑁
𝑆𝑢 = 𝑢𝑖 𝑆𝑢𝑢 = 𝑢𝑖2 𝑆𝑦 = 𝑌𝑖 𝑆𝑢𝑦 = 𝑢𝑖 𝑌𝑖
𝑖=1 𝑖=1 𝑖=1 𝑖=1
NonLinear Regression

𝑦 𝑡 = 𝐾𝑀(1 − 𝑒 −𝑡/𝜏 )

Linearization
Chapter 7

𝑦 𝑡 𝑡
ln 1 − =− (7−13)
𝐾𝑀 𝜏

• The slope is given by:


1
𝑠𝑙𝑜𝑝𝑒 = −
τ

As an alternative to nonlinear regression, graphical method can


be used quickly to find approximate values of time constant
Graphical Method

• Simple transfer function models can be obtained graphically from


step response data.
• A plot of the output response of a process to a step change in input
is sometimes referred to as a process reaction curve.

Process reaction curve - The simplest and most often used


method. Gives nice visual interpretation as well.
The PRC method involves 4 actions:
1. Allow the process to reach steady state.
2. Introduce a single step change in the input variable.
3. Collect input and output response data until the process again
reaches steady state.
4. Perform the graphical process reaction curve calculations.
First Order Models

𝑌 𝑠 𝐾
• The response of a first-order model, = , to
𝑋 𝑠 𝜏𝑠+1
a step change of magnitude M is:

𝑦 𝑡 = 𝐾𝑀(1 − 𝑒 −𝑡/𝜏 )

• The initial slope is given by:


𝑑 𝑦 1
= (7−15)
𝑑𝑡 𝐾𝑀 𝑡=0 τ

• The gain can be calculated from the steady-state changes


in u and y:
Δ𝑦 Δ𝑦
𝐾= =
Δ𝑢 𝑀
First Order Models

Figure 7.3 Step response of a first-order system and graphical


constructions used to estimate the time constant, τ.
Example 1
■ Figure below shows the response of the temperature T in a CSTR
to a step change in feed flow rate w from 120 to 125 kg/min.
Find an approximate first-order model for the process and these
operating conditions.

Δ𝑇 160 − 140
𝐾= =
Δ𝑤 125 − 120
𝑜𝐶
=4
𝑘𝑔/𝑚𝑖𝑛

τ = 5 min

𝑇′ 𝑠 4
=
𝑤′ 𝑠 5𝑠 + 1
Exercise
■ A single –tank process has been operating for a long period of time
with the inlet flow rate q𝑖 = 30.4 𝑓𝑡3/𝑚𝑖𝑛. After the operator
increase the flow rate suddenly by 10%, the liquid level in the tank
changes as shown in below table. Assuming that the process
dynamics can be described by a first order model, calculate the
steady state gain and time constant using 2 graphical method:

a) From the time required for


Time, Height, Time, Height,
min (𝒖𝒊 ) ft (𝑌𝑖 ) min (𝒖𝒊 ) ft (𝑌𝑖 )
the output to reach 63.2%
of the total change.
0.0 5.50 1.4 6.37
b) From the initial slope of
0.2 5.75 1.6 6.40
the response curve
0.4 5.93 1.8 6.43
0.6 6.07 2.0 6.45
0.8 6.18 3.0 6.50
1.0 6.26 4.0 6.51
1.2 6.32 5.0 6.52
Solution
6.60

6.40

6.20

0.632
6.00

5.80

5.60

5.40
0.0 1.0 2.0 3.0 4.0 5.0 6.0
τ
Solution
6.60
slope
6.40

6.20

6.00

5.80

5.60

5.40
0.0 1.0 2.0 Slope
3.0 at t=0
4.0 5.0 6.0
Drawback of First order model:

■ The true process model neither first order nor linear.


Only the simplest processes exhibit such ideal dynamics.
■ The output data are usually corrupted with noise.
■ Another process input (disturbance) may change during
the duration of step test.
■ It can be difficult to generate a perfect step input

In order to account for high order dynamics, a time delay


term can be included. This modification can improve the
agreement between model and experimental response.
END

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy