Texas A&M Uses NI LabVIEW Control Design Toolkit and LabVIEW Control Design and Simulation Module for Novel PID Control Design Technique
"This will effectively change the design process from model-based to measurement-based control design."
- Shankar Bhattacharyya,
Texas A&M University
Developing computer-aided design tools for fixed-order and proportional integral derivative (PID) controllers with guaranteed stability, performance, and power.
Using NI LabVIEW, the LabVIEW Control Design Toolkit, and the LabVIEW Control Design and Simulation Module to design, analyze, and simulate fixed-order and PID controllers that meet all of the control design objectives.
Shankar Bhattacharyya - Texas A&M University
Ubiquity and Challenges of PID Controllers in the Industry
Fixed-order controllers, particularly PID controllers, are widely used in aerospace, motion control, process control, manufacturing, robotics, and disk drives. Engineers implement current designs using ad hoc tuning rules that have been developed based on empirical observations and industrial experience. This approach is time-consuming and expensive to use during initial tuning or retuning if plant characteristics change. Also, the engineer may choose gains that result in a controller that is only marginally stable.
Developing an Approach Using the NI LabVIEW Control Design and Simulation Module
We developed computer-aided design tools for fixed-order and PID controllers with guaranteed stability, performance, and power. Our current research focuses on straightforward algorithms using LabVIEW to help guide the industrial practitioner in the control design process given certain assumptions about the plant model. The model can be defined as a transfer function. Thus far, we have considered the following types of plants:
- Continuous-time plants of arbitrary order without time delay
- Discrete-time plants of arbitrary order
- Continuous-time plants of first order with time delay
We used native LabVIEW to develop the algorithms for our customized control design process. We also used the LabVIEW Control Design Toolkit to construct the controller and plant models and to analyze time and frequency response parameters of the resulting closed-loop system. To simulate the closed-loop system dynamics, we implemented the LabVIEW Control Design and Simulation Module.
We applied our theoretical development to PID controller design using LabVIEW. Given a plant model, we can calculate stabilizing proportional, integral, and derivative controller gain sets using nested linear programming techniques. Then we display all of the resulting gain sets using a three-dimensional graph. The user can view the entire allowable design space with this graph. We guide the user through the rest of the design process based on the computed gain sets. After choosing a stabilizing proportional gain from the allowed range of proportional gains, the control designer can point and click anywhere within a closed curve on a two-dimensional graph to select allowable derivative and integral gains. The user can observe the time and frequency response characteristics of resulting closed-loop systems and iterate, as needed, to choose gain values that best suit the design constraints.
In addition, we used the LabVIEW Control Design Toolkit to construct controller and plant models in transfer function form. We also used the time and frequency response analysis tools in the toolkit to calculate closed-loop characteristics such as settling time, rise time, percent overshoot, and gain and phase margins.
Using the LabVIEW Control Design and Simulation Module, we computed and plotted the impulse response of the closed-loop system. We wired plant parameters and controller gains from the design process directly into a LabVIEW simulation diagram within the same VI. We constructed the controller model using linear integrator and derivative elements along with summation and multiplication signal arithmetic within the simulation diagram. Also, we constructed the plant model with the continuous linear transfer function block and modeled time delay using the transport delay block.
For the simulation ODE solver, we chose the Runge-Kutta 45 variable step-size solver and we increased computational efficiency while staying within per-step error tolerances.
Benefits of Using LabVIEW for Control Design and Simulation
Using LabVIEW, we quickly implemented the algorithms and logic for our control design process. We also developed a unique GUI in which the industrial practitioner can make design decisions and then gain an immediate understanding of the resulting control system dynamics. The fixed-order and PID controller designs have improved stability, performance, and power and the overall design cycle is shortened.
The LabVIEW Control Design and Simulation Module can be used directly with LabVIEW and the LabVIEW Control Design Toolkit to combine graphical simulation with graphical programming and analysis. Users can implement the entire design and simulation process in one environment with a flexible and customizable user interface.
We will begin the control design process with measured frequency-response functions for the plant rather than a transfer function model of the plant to effectively change the design process from model-based to measurement-based control design. This development will significantly reduce the time and difficulty of the overall design process because we will be able to bypass the modeling and system identification steps needed to derive the transfer function plant model.
Bookmark and Share
Explore the NI Developer Community
Discover and collaborate on the latest example code and tutorials with a worldwide community of engineers and scientists.
Who is National Instruments?
National Instruments provides a graphical system design platform for test, control, and embedded design applications that is transforming the way engineers and scientists design, prototype, and deploy systems.