Scribd
Upload
164 views2 pages

DC Gain Calculation with Integration

This document discusses finding the DC gain of systems with an integration. It explains that for continuous and discrete systems, if there is an integration term, the DC gain will be infinity. However, the ratio between the steady-state input and steady-state derivative of the output can be used instead. This involves differentiating the output function. The document provides an example of finding the "derivative DC gain" for both a continuous transfer function and its corresponding discrete transfer function. It shows how to calculate the derivative gain and ensure the discrete system matches the continuous system.

Uploaded by

rosita61
Copyright
© Attribution Non-Commercial (BY-NC)
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
164 views2 pages

DC Gain Calculation with Integration

This document discusses finding the DC gain of systems with an integration. It explains that for continuous and discrete systems, if there is an integration term, the DC gain will be infinity. However, the ratio between the steady-state input and steady-state derivative of the output can be used instead. This involves differentiating the output function. The document provides an example of finding the "derivative DC gain" for both a continuous transfer function and its corresponding discrete transfer function. It shows how to calculate the derivative gain and ensure the discrete system matches the continuous system.

Uploaded by

rosita61
Copyright
© Attribution Non-Commercial (BY-NC)
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

DC Gain w/Integration

February 7, 2008

DC Gain of Systems with an Integration

The Problem

There are situationslike my unfortunate choice of example in class on Wednesday for a demonstration of the PoleZero Mapping Methodin which one is faced with nding the DC gain of a system possessing an integration. The DC gain is that ratio of the steady-state output to the steady-state input, and a system possessing an integration has a steady-state output which is a ramp there is no steady-state value.

1.1

Finding the DC Gain

Consider both continuous and discrete LTI systems: Y (z ) Y (s) = G(s) or = G(z ) U (s) U (z ) (1)

Finding the DC gain of an LTI system (either continuous or discrete) is based on the Final Value Theorem, and in either case involves the following formul: Continuous System: Discrete System: y () = G(s)s=0 = G(0) u() y () = G(z )z=1 = G(1) u() (2) (3)

In both cases, if the system has an integration (free s term or z 1 term in the denominator) the result will be .

1.2

One SolutionDierentiate!

There is a solution: nd the ratio between the steady-state input and the steady-state derivative of the output. This requires one to dierentiate the output. This is similar in both the continuous and discrete domains, but a little dierent... 1.2.1 Dierentiation in the continuous domain

In the continuous, or s domain, dierentiation is done by simply multiplying by s, that is, referring to (1): (s) Y = sG(s) U (s) (s)this just represents the Laplace transform of y Please excuse the bastard notation of Y (t). 1.2.2 Dierentiation in the discrete domain (4)

The derivative in the discrete domain can be done by a rst dierence: y k = y k y k 1 T =


1 1 z1 (z ) = Y (z ) z Y (z ) = Y (z ) 1 z Y = Y (z ) T T Tz

(5)

Thus to dierentiate in the discrete domain, simply multiply by

z1 . Tz

DC Gain w/Integration

February 7, 2008

2
2.1

An Example
Continuous System

Consider the continuous transfer function G(s) = 5 s(s + 10) (6)

To nd the DC gain (steady-state gain) of this transfer function between the derivative of the output and the steadystate input, multiply by s, then let s 0. Thus we have derivative DC gain of G(s) = lim sG(s) = 0.5
s=0

(7)

2.2

Discrete System

Lets use the Pole-Zero Mapping Method to nd a G(z ) as I tried in class. Use sampling period T = 0.02 seconds. Mapping both poles using z = eT s and placing two zeros at z = 1, we have G(z ) = (z + 1)2 (z 1)(z 0.8187) (8)

But we need to make the DC gain of (8) match the DC gain of (6). We can still do this with the derivative gain, which we found in (7). Applying the result in (5) to this G(z ), we get derivative DC gain of G(z ) = lim z1 (z + 1)2 = 1103.1 T (z 1)(z 0.8187) (9)

z =1

To make the DC gain of G(z ) match that of G(s) we must apply a numerator constant which is equal to 0.5/1103.1. This results in nal result of G(z )P Z = (4.5325e 04)(z + 1)2 (z 1)(z 0.8187) (10)

2.3

Validation

I would encourage you to nd the step response of both the G(s) and G(z ) of (6) and (10) to verify that they are similar.

You might also like