Mingxiao Lu - Xi’an Jiaotong University
Zhou Jing - Xi’an Jiaotong University
Zheng Xu - National Instruments
The Department of Automation Science and Technology in the College of Electronic and Information Engineering manages the Automatic Control Principles Laboratory at Xi’an Jiaotong University. The automatic control principles class and its corresponding lab course are the most important specialized fundamental courses for students majoring in automatic control. There is also a separate lab course in automatic control principles offered to students majoring in other engineering disciplines. The goal of these two laboratory sessions is not only to help students understand the basic theories and concepts introduced in lecture, but also to provide an intuitive perception of how engineers implement these theories in practice.
In traditional automatic control experiments, students use preconstructed experiments to verify or observe control principles, such as time or frequency domain characteristics of first- or second-order systems, by simply following the step-by-step guidance in a lab textbook. Using this method, even without sufficient preparation or comprehension, students can finish the experiments as long as they follow the steps in the textbook and get the expected results. A scripted lab session like this leaves little room for students to take initiative or explore. As a result, they do not get hands-on experience or develop their problem solving skills, both of which are critical for engineering students.
We wanted to restructure the teaching lab and replace the traditional verification-oriented experiment setup for control students with design-oriented experiments. This method encourages students to use proactive thinking and it develops their ability to solve practical engineering problems through hands-on, experimental teaching. For students from other majors, we expected that changing the verification-oriented experiments would inspire them to study and engage with controls concepts.
Furthermore, by incorporating real-world engineering into both labs, students gain a more intuitive understanding of the core concepts taught in the class and can better apply those concepts.
After researching options, we decided on the NI ELVIS platform and LabVIEW software for our updated automatic control principles teaching laboratory. The deciding factors included the ability to integrate the features of the many measurement devices students were previously using into one form factor with NI ELVIS. Coupled with LabVIEW, we could also facilitate the development of additional experiment features.
NI ELVIS and LabVIEW replace step-by-step verification experiments with a hands-on lab environment for the students. Additionally, we can expand NI ELVIS through add-on boards from third-parties such as Quanser, and teachers can even design their own boards for experiments.
So far, we have developed several innovative experiments based on the NI platform. Different combinations of these experiments are available for different majors and credit hours. Students majoring in automatic control need to finish several verification experiments such as time domain response analysis, frequency domain analysis, stability analysis, and lead compensation, as well as one design experiment focused on proportional integral derivative (PID) control of motor speed.
Even in the verification experiments, students engage with the engineering design process by creating the first- and second-order systems that need to be analyzed on the NI ELVIS platform using operational amplifiers, resistors, capacitors, and other elements. They then observe and measure the impacts on the performance of these systems by changing parameters in the LabVIEW programs provided by the teachers. This offers students a deeper understanding of the system characteristics and also incorporates the knowledge learned from previous courses such as analog electric circuits.
We put students into groups of three during the design experiments. Each group needs to independently design a DC motor PID speed control system. The teacher provides motors as the control plant. Students need to build the circuitry on the NI ELVIS platform to interface with an optocoupler switch speed measurement device to measure motor speed. They then use the NI ELVIS programmable power supply to output different voltages according to the current speed to drive the motor to match the speed setpoint. Software development involves acquiring the output pulse signals of the optocoupler switch, calculating rotational speed, calling the PID algorithm provided by LabVIEW, setting proper PID parameters, and controlling the ouput voltage of the NI ELVIS.
The junior students in the fall semester had not been provided with LabVIEW courses yet, so we provided separate code snippets as a reference. The students still needed to integrate these snippets into a complete application and independently program their core control algorithm. LabVIEW is an intuitive graphical programming language that provides off-the-shelf functions such as PID control. This design experiment is challenging for junior students, but they can succeed with appropriate effort.
Through these hands-on design experiments, students can gain a deeper understanding of related control theories as well as their practical applications. Meanwhile, their practical skills, innovation, problem-solving capabilities, and teamwork all improve during the construction of a complete system.
In the conclusion of one experiment report, students said, “This experiment gave us a more in-depth understanding and mastering of automatic control, especially the impact on characteristics by adjusting each PID parameter. We were able to remember these very well through this lab session. With its graphical programming language and block diagram output, LabVIEW provides an extremely convenient programming method for designers and analysts, making program design much easier and analysis simpler.”
Because of time constraints, verification experiments remain the focus for the automatic control principles course for students not majoring in automatic control. However, we expect better results through updated controls experiment content. Several experiments we designed use the QNET DC motor teaching board provided by Quanser as the control plant, together with the NI ELVIS platform and LabVIEW software. Students do not need to write programs or create circuits themselves, but they still perform a series of hands-on experiments to model a DC motor system, calculate PID constants derived from the model, and control speed and position of the DC motor. This laboratory curriculum still gives them the opportunity to closely relate control theories to engineering practices.
The automatic control principles lab has been successful since we restructured it in 2011. Due to this success, we established a new system for teaching the automatic control principles lab based on the NI platform.
In 2013, the lab expanded from eight NI ELVIS stations to 21. Professors have already compiled the lab teaching textbook, which will be published in early 2014. In the future, purchasing additional stations will help more students do experiments in one lab session as professors continue to develop and evolve the lab’s content.