Echelon LonPoint Application and Plug-In - Manual (Page 96)

9-12 PID Controller Functional Block
failure conditions are determined by the functional
blocks that contain the corresponding input network
variables.
Status
This tab allows you to view and change the status of a PID controller functional
block. See Status in Chapter 2 for more information.
LonPoint PID Controller Tuning
This section describes a simple method that can be used to set initial values for the
PID coefficients. An open-loop experiment is done to determine the static gain and
the basic dynamic properties of the process being controlled. These values are
entered into a program that calculates the P, I, and D coefficients for the PID
controller. The coefficients are loaded into the PID controller using the LonPoint
Plug-in and the LonMaker for Windows Integration Tool. The coefficients match or
tune the controller for the controlled process.
Figure 8.1 shows an example PID loop for reference. The Analog Inputs come in
from the left and connect to the PID setpoint (SP)and process variable (PV) inputs.
AI- 2 controls the setpoint, which can range from zero to one hundred percent. In
practice the setpoint and process variable will be in terms of some engineering unit,
temperature in degrees C for example. The PID calculates the control value, CV,
and sends it to the Analog Output, AO- 1. AO- 1 is wired to an actuator that drives
the plant or process. The process is sampled by the Analog Input, AI- 1, and
connecting this back to the PV input of the controller closes the loop.
In this example there are unconnected inputs on the PID. When inputs are not
connected, default inputs values are used. These defaults can be set on the Input
Defaults tab.
Figure 8.1 PID Example Drawing
To calculate the PID coefficients, you can perform an experiment. To do the
tuning experiment, put the PID controller into manual mode, set the manual value
and wait for the sensed process variable to reach equilibrium. Then the manual
value is changed to a different value; this is the ‘step’ part of the experiment. The
proper step size depends on the controlled process. With some trial and error
experience, it should not be difficult to choose a reasonable step size. The largest