Linear System
In any system, if there exists a linear relationship between two variables, then it is said that it is a linear system.
For example, the equation
represents a linear system. It means that if K is constant then the relationship (1.1) represents a linear relationship between two variables y and x. In general, any governing differential equations between two variables x and y in the form of
is linear, where n and m represent the order of differential equations, and a n , b m are constants. For real system n>m, any other form of equations that is not similar to equation () is called nonlinear system.
There are extensive theories that deal with linear systems, but the theories on nonlinear systems are very complex and little.
Example 1
The circuit diagram of equivalent DC servo motors is shown in Fig..
Fig. 1.1
Equivalent circuit diagram of a DC servo motor
The governing differential equation may be written as
where V i , I , m are the input voltage, current, and angular speed. R and L are the resistance and inductance, respectively. This represents a linear system, where m is the output variable and V i represents the input voltage.
For DC servo motor, we can write
where K i , J are the torque constant and rotor moment of inertia.
Eliminating T , from equations () yields
Equation (), we ignore the external torque acting on the motor. If we consider the external torque, the governing differential equation would have two input variables and one output variable.
For linear systems, the principle of superposition holds. It means that if input x1 causes output y1 and input x2 causes output y2, then input x1+x2 causes output y1+y2. This is a powerful principle, and we will use it throughout this book.
Nonlinear Systems
There are different kinds of nonlinearities. For example, onoff control systems are inherently nonlinear. Transport lag, saturation, and transport lag are other kinds of nonlinearities. These kinds of nonlinearities cannot be solved with linear control theory. This is shown in Fig..
Fig. 1.2
Some discontinuous nonlinearities
For linearized equation, it is better to use Laplace Transform. In this way, the differential equations become algebraic equation in s . Throughout this book, the lower case s represents Laplace Transform.
Some nonlinearity is continuous, and they can be solved by the linearization technique. One example of this kind of nonlinearity is
This is shown in Fig.
Fig. 1.3
A continuous nonlinearity
Linearization Technique
If there is a continuous nonlinearity in the form of
Assuming small perturbation from the equilibrium point, equation () can be linearized as
or it can be written as
In equations () can be written as
where
K is constant at an operating point. Throughout this book, the lower case variable represents small perturbation from equilibrium point. This is shown in Fig..
Equation () represents one variable system. For a multivariable system, similar linearized equation can be obtained.
The solution of the governing equation simplifies if Laplace Transform is used.
Laplace Transform
By the definition, the Laplace Transform is defined as
By taking the Laplace Transform, the variable t is eliminated and the result is only function of s .
Equation () appears to be very complicated, and indeed for complicated transformation, the integral becomes very complex. Fortunately, for control systems only a few functions are needed.
This is a simple integration, and the integral becomes