Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
How can I use the PID auto tune example in my process?
rhray
Member
07-25-2001 06:27 AM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
Also in the example I notice the output is alway 20 less then the setpoint. Plus I don't understand the [PV] expression used as the PV. Can you explain
Greg Caesar
Member
08-06-2001 11:35 AM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
The Lookout PID Auto Tune Example is designed to allow you to connect variables from your "real" process to various items in this example. This allows you to tune your feedback control loops online with this example.
The example uses a software simulation of a system by default. This is so that when the example is opened, it demonstrates the PID control loop and autotuning functionality immediately, without any changes to connections. It turns out that this "simulated process" was mathematically defined in such a way that it just so happens the output stabilizes to 20 less than the setpoint. If you change the example to connect to a real system, or change the math in the "temperature_simulated" expression, then this will not be the case.
The [PV] expression is a static symbolic link. This allows you to connect the PV (Process Variable) expression to any data member in any running process on any computer on your network. Since by default this example uses a software-simulated system for reasons explained above, this expression connects to "Temperature_simulated" by default.
To change the connections within this process from the default, built-in simulation, simply follow the step-by-step procedures in the documentation included with the example. The procedure has you connect the PV expression to your process variable, and the POTVoltage expression to your output or manipulated variable. It also walks you through re-scaling of all the outputs, as the real-world values are likely engineering units and are probably not scaled from 0 to 100.
As a final note, this auto-tuning procedure uses a Ziegler-Nichols procedure, which is an industry-standard procedure for tuning feedback control loops. Personally, I think it results in too much derivitive action (the "Rate" constant), which causes your manipulated variable to jitter when small amounts of system noise are present. So I usually reduce or set to zero the "Rate" constant after using this Auto-Tune procedure, (ie PI control vs PID control).
And as always, watch your control loops closely and be ready to go to manual if necessary when you are testing new tuning constants.
Regards,
Greg Caesar
National Instruments,
Applications Engineer
The example uses a software simulation of a system by default. This is so that when the example is opened, it demonstrates the PID control loop and autotuning functionality immediately, without any changes to connections. It turns out that this "simulated process" was mathematically defined in such a way that it just so happens the output stabilizes to 20 less than the setpoint. If you change the example to connect to a real system, or change the math in the "temperature_simulated" expression, then this will not be the case.
The [PV] expression is a static symbolic link. This allows you to connect the PV (Process Variable) expression to any data member in any running process on any computer on your network. Since by default this example uses a software-simulated system for reasons explained above, this expression connects to "Temperature_simulated" by default.
To change the connections within this process from the default, built-in simulation, simply follow the step-by-step procedures in the documentation included with the example. The procedure has you connect the PV expression to your process variable, and the POTVoltage expression to your output or manipulated variable. It also walks you through re-scaling of all the outputs, as the real-world values are likely engineering units and are probably not scaled from 0 to 100.
As a final note, this auto-tuning procedure uses a Ziegler-Nichols procedure, which is an industry-standard procedure for tuning feedback control loops. Personally, I think it results in too much derivitive action (the "Rate" constant), which causes your manipulated variable to jitter when small amounts of system noise are present. So I usually reduce or set to zero the "Rate" constant after using this Auto-Tune procedure, (ie PI control vs PID control).
And as always, watch your control loops closely and be ready to go to manual if necessary when you are testing new tuning constants.
Regards,
Greg Caesar
National Instruments,
Applications Engineer
