rruiter87 / TcControl

TwinCAT library for PID control and signal filtering

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

TcControl

TwinCAT library for PID control and signal filtering.

Examples

Signal

The basic unit of this library is the Signal function block. The most basic usage would be the following where a new value is added and can be retrieved.

PROGRAM MAIN
VAR
    signal : Signal;
    value : LREAL;
END_VAR

signal.Update(value:=5); // Add new value to signal
value := signal.Out; // Retrieve the current value, in this case it would be 5

The power of signal becomes clear when combined with .Differentiate

PROGRAM MAIN
VAR
    signal : Signal;
    value : LREAL;
END_VAR

signal.Update(value:=1).Differentiate(deltaTime:=1);
value := signal.Update(value:=2).Differentiate(deltaTime:=1).Out; // 1

Or .Iir

PROGRAM MAIN
VAR
    signal : Signal;
    value : LREAL;
END_VAR

signal.Update(value:=1.4).Iir(decay:=0.5);
value := signal.Update(value:=8.6).Iir(decay:=0.5).Out; // 4.65

Or combining both

PROGRAM MAIN
VAR
    signal : Signal;
END_VAR

signal.Update(value:=7).Differentiate(deltaTime:=1).Iir(decay:=0.5); // singal.Out = 0
signal.Update(value:=8).Differentiate(deltaTime:=1).Iir(decay:=0.5); // signal.Out = 0.5
signal.Update(value:=9).Differentiate(deltaTime:=1).Iir(decay:=0.5); // signal.Out = 0.75

Finally the differentiated/filtered signal can be put into a controller. Currently the library only contains the simple PidController

PROGRAM MAIN
VAR
    signal : Signal;
    parameters : PidParameters := (Kp:=2, Ki:=0.25, Kd:=0.5);
    pid : PidController;
END_VAR

pid(parameters:=parameters, cycleTime:=1);
signal.Controller := pid;
signal.Update(3).Control(setpoint:=4); // signal.Out = 2.75
signal.Update(3.5).Control(setpoint:=4); // signal.Out = 1.125

See the unit tests Singal_Tests for more examples.

Custom controller

You can define your own controller, as long as it implements the IController interface. Then you pass it to your signal instance via signal.Controller := customController;.

Other function blocks

The function blocks used in Signal can also be used as stand-alone function blocks. The following function blocks are defined.

  • Derivative: Take derivative of a value.
  • InfiniteImpulseResponse: Infinite impulse response (IIR) filter.
  • Pid: PID controller
    • Pid.DifferentialPartLimit: Limit the differential part. Convenient when the output is calculated for the first time. The limit prevents it from having a very large value after the first call.
    • Pid.PreviousIntegralPart: Set previous integral part on first call. Prevents integrator term for needing to build up.

Derivative

PROGRAM MAIN
VAR
    derivative : Derivative;
END_VAR

derivative(value:=1, deltatime:=1); // derivative.Out = 0 (derivative is always 0 on the initial call)
derivative(value:=2, deltatime:=1); // derivative.Out = 1

See Derivative_Tests for more examples.

InfiniteImpulseResponse

PROGRAM MAIN
VAR
    iir : InfiniteImpulseResponse;
END_VAR

iir(new:=5, decay:=0.95); // iir.Out = 0.25
iir(new:=5, decay:=0.95); // iir.Out = 0.4875
iir(new:=5, decay:=0.95); // iir.Out = 0.71313

See InfiniteImpulseResponse_Tests for more examples.

PidController

PROGRAM MAIN
VAR
    pid : PidController;
    parameters : PidParameters := (Kp:=0.1, Ki:=0, Kd:=0.01);
END_VAR

// limits the difference on the initial call
pid.DifferentialPartLimit := 3;
// pass the pid parameters (only need to be done once, unless you change them)
pid(parameters:=parameters, cycleTime:=1);
// Cyclicly call Update
pid.Update(setpoint:=10, actual:=6); // pid.Out = 0.4
pid.Update(setpoint:=10, actual:=8); // pid.Out = 0.18

About

TwinCAT library for PID control and signal filtering

License:MIT License


Languages

Language:Python 100.0%