Analysis
LabVIEW add-on: PID Control
Overview
The PID Control Toolkit add sophisticated control algorithms to your instrumentation software development system. By combining the PID and fuzzy logic control functions in this toolkit with the math and logic functions in LabVIEW, you can quickly develop programs for automated control.
PID Tools
The PID functions implement a wide range of PID algorithms with error-squared and external-reset feedback. They also implement lead-lag compensation and setpoint ramp generation. Control strategies include multiloop cascade, feedforward, minimum and maximum override, and ratio/bias. The PID algorithms feature bumpless auto/manual transfer, antireset windup, direct/inverse action, manual output adjustment, and a run/hold switch.
Control Strategy Design
You can design PID control strategies, scale I/O values from engineering units to percentages, and set up timing of the PID algorithms. Finally, you can use tuning procedures for both Closed-Loop (Ultimate Gain) and Open-Loop (Step Test).
Autotuning
Autotuning can be used to automatically improve the performance of your stable P, PI, or PID controller. The autotuning algorithm is a setpoint relay feedback method that automatically tests the control system to determine new PID parameters that will improve the performance of the controller with respect to parameters such as risetime and overshoot. This method maintains closed-loop control during the tuning process, therefore maintaining stability of the system controlled. The PID Tools also feature gain scheduling. You can improve performance in defining several regions of operation for the controller.
PID Control Examples
Examples for process control included in the PID Control Toolkit are:
- Simulation VIs
- Tank Level
- General PID
- Plant Simulator
- Cascade and Selector
- PID with MIO Board
- Lead-Lag
Fuzzy Logic Tools
Fuzzy logic can be used to accelerate development of control for nonlinear or highly complex systems. It is easy to comprehend because the control strategy is implemented with simple, intuitive, linguistic rules. Thus, the need for highly complex and difficult-to-understand control strategies is eliminated. In addition to control applications, the fuzzy logic software can be used for expert decision making, such as pattern recognition or fault diagnosis. In addition to helping you design your control system, this point-and-click fuzzy logic software includes functions for implementing your fuzzy control system in LabVIEW.
Fuzzy Logic Control Designer
Each fuzzy controller can have up to four inputs and one output. For systems with large numbers of controller inputs, you can cascade multiple fuzzy controllers for control while implementing rules that are easy to understand. Linguistic terms for rules are defined with membership functions of the following shapes - triangular, trapezoidal, Z-shaped, S-shaped, or singleton. Rules are generated automatically to create a complete rule base of all possible combinations of inputs using the interactive fuzzy logic software. The output of each rule is manually selected and weight values may be assigned for the purpose of controller tuning. Defuzzification methods include Center of Area, Center of Maximum, and Mean of Maximum. All designed controller information is stored in a data file for implementation in the application.
Controller VIs
This toolkit includes VIs to be used in your application for control or decision making. All controller information from the fuzzy logic control designer is saved to a data file and is retrieved and grouped into a single cluster with the Load Fuzzy Controller VI. This cluster of information is wired in your application to the Fuzzy Controller VI. Controller inputs are wired into this VI and the controller output is calculated, based on your designed controller, and returned to your application. You can easily integrate fuzzy logic with NI-DAQ and data acquisition hardware. Examples provided with the software include fuzzy control examples, a pattern recognition example, and an example of implementing a fuzzy controller with DAQ functions.
Membership Functions
Triangular
Trapezoidal
Z-shaped
S-shaped
Singleton
Defuzzification Methods
Center of Area
Center of Maximum
Mean of Maximum
Related Application Note
Pricing and Purchase Information
