How can a function be analytic in the s-plane

  • Thread starter Thread starter mech-eng
  • Start date Start date
  • Tags Tags
    Function
AI Thread Summary
A function is considered analytic in the s-plane if it possesses a complex derivative throughout the complex plane, except at specific points where it may be undefined, such as the zeros of the denominator in a rational function. Analytic functions, including all polynomials, have significant properties such as the existence of higher-order derivatives and convergent power series expansions. The discussion highlights the importance of understanding complex analysis for automatic control systems, suggesting that foundational knowledge in calculus is necessary before delving deeper into complex analysis. Numeric and stochastic methods are employed when differential equations lack analytic solutions, with numerical methods involving computer simulations and stochastic methods addressing uncertainty in models. Overall, a solid grasp of these concepts is essential for further studies in control theory.
mech-eng
Messages
825
Reaction score
13
analitic func.png
Hi, I try to learn the subject of automatic control and there is an explanation which I cannot understand, probably because of background of mathematics. Here is the explanation. I would like to ask that how a function be analitc in the s-plane and would you like to explain it.

Source: Automatic Control Systems by Kuo

Thank you
 
Engineering news on Phys.org
mech-eng said:
View attachment 98948 Hi, I try to learn the subject of automatic control and there is an explanation which I cannot understand, probably because of background of mathematics. Here is the explanation. I would like to ask that how a function be analitc in the s-plane and would you like to explain it.

Source: Automatic Control Systems by Kuo

Thank you
Here is an article on the properties of analytic functions:

https://en.wikipedia.org/wiki/Analytic_function
 
I think analytic would be opposed to not solvable analytically, i.e. numeric or stochastic methods were used to arrive at a solution.
 
@jeff Resenbury
Can you please explain me what numeric or stochastic methods are used actually to solve it.
 
Jeff Rosenbury said:
I think analytic would be opposed to not solvable analytically, i.e. numeric or stochastic methods were used to arrive at a solution.
In this context "analytic" means that it has a complex derivative in the complex plane, s. All polynomials are analytic in the entire complex plane and the quotient of polynomials is analytic everywhere in the complex plane except at the zeros of the denominator. So the function G(s) in the OP is analytic everywhere for s in the complex plane except at s = 0, -1, and -3.
 
FactChecker said:
In this context "analytic" means that it has a complex derivative in the complex plane, s. All polynomials are analytic in the entire complex plane and the quotient of polynomials is analytic everywhere in the complex plane except at the zeros of the denominator. So the function G(s) in the OP is analytic everywhere for s in the complex plane except at s = 0, -1, and -3.

What does complex derivative refer to?

Thank you.
 
mech-eng said:
What does complex derivative refer to?

Thank you.
The derivative of a complex-valued function, f, of a complex variable, z, at point z0 is the complex number d, such that d = limz->z0( f(z)-f(z0 ) / (z-z0). The limit value, d, must be the same no matter what direction z approaches z0 from.

When a function, f, has this property at each point in an area, the consequences are profound: If it has one derivative, then it also has all higher order derivatives. It has a convergent power series expansion. It's integral also has very special properties. I am sure that you will see a lot of this as you continue your studies.
 
Last edited:
FactChecker said:
The derivative of a complex-valued function, f, of a complex variable, z, at point z0 is the complex number d, such that d = limz->z0( f(z)-f(z0 ) / (z-z0). The limit value, d, must be the same no matter what direction z approaches z0 from.

When a function, f, has this property, the consequences are profound. If it has one derivative, then is has all higher order derivatives. It has a convergent power series expansion. It's integral also has very special properties. I am sure that you will see a lot of this as you continue your studies.

Is this topic belong to Complex Analysis Theory? In which books and at which titles of those books I can find this topic? Even I do not know enough Real Analysis [Calculus], first should I learn Real Analysis?

Thank you.
 
mech-eng said:
Is this topic belong to Complex Analysis Theory? In which books and at which titles of those books I can find this topic? Even I do not know enough Real Analysis [Calculus], first should I learn Real Analysis?

Thank you.
Yes. It is complex analysis. Real analysis would emphasize things that will not help much. I would look for a book in control laws that will summarize the basic complex analysis that is needed to understand control laws. If you need more, you can just read sections of complex analysis books. And the complex analysis books should be for engineers, not mathematicians.

PS. If you mean calculus when you say "Real Analysis", you do need some understanding of introductory undergraduate calculus. A Real Analysis book is very different.
 
Last edited:
  • #10
StephenBarton said:
Can you please explain me what numeric or stochastic methods are used actually to solve it.
Not all differential equations have analytic solutions. When these problems need to be solved they are sometimes solved either numerically or stochastically.

To solve numerically, use a computer to solve small bits as if they were differential elements. Make sure to run an error analysis. An example would be weather modelling.

Stochastic modelling is sometimes done when the certainty of the model is in question. Insurance companies use them to set rates for example. Something like an accident rate might have a function found by curve fitting, but one can never tell if that function truly represents the data.

The point is that not all functions are "nice" for some value of nice. The definitions need to be abandoned or at least modified when a function is weird enough to not play by normal rules.
 
Back
Top