All of the data presented in this page was obtained using the selected 18x8E propeller and the final electric components listed at the end of the the Propulsion System’s page.
Prior to testing the P.I Controller at the desired set point (925W-950W), as preliminary test, I let the set point be 500W to determine how well it performs under limits below its intended application.
500W Set Point Test
The figure above shows the power data collected over a 50 second interval with the integration of the P.I controller at a 500W set point. We can see that the average power (red curve) being utilized by the system levels at around 500W, which is what I expected. However, you cannot see how even though the throttle signal was being increased the power consumed by the propulsion system remained around the set point. Unfortunately at the time this data was being recorded, I forgot to record the change in the width of the throttle signal. I think showing the direct relationship between the power consumption and the control signal would have made the figure above more meaningful.
In order to get a better understanding of the functionality of the P.I controller the following plots will provide detail about the throttle signal as well.
SAE Power Limiter Test
The figure above shows the power and throttle data collected over a 40 second interval without the integration of the P.I controller. We can see that the power being utilized by the system drops significantly once the average power (red curve) exceeds the 1kW limit. The substantial reduction in power seen in this figure is the numerical representation of the motor stalling, which can potentially end our participation in the competition if it were to occur while airborne. The width of the control signal (PWM signal) that caused the limiter to engage was determined to be approximately 1.73ms.
The P.I Controller is designed to be a pre-limiting device with the overall goal of preventing the SAE Limiter from taking action.
P.I Controller Test
The figure above shows the power and throttle data collected over a 45 second interval with the integration of the P.I controller at a 950W set point. We can see that the average power (red curve) being utilized by the system never exceeds the 1kW limit regardless of what the control signal input is. In the SAE Limiter Test figure we saw that the limiter engaged at a throttle signal width of approximately 1.73ms, but in this figure the throttle signal width exceeds the 1.73ms “max” control signal without causing any reductions in power consumption. The average power utilized during this test run while the throttle was fully engaged was about 915W, which is not the 950W that I expected, but I suspect that this discrepancy can be fixed with the proper Ki and Kp (integral and proportional constants) values. The proportional and integral constants used for all P.I control examples were 0.7 and 0.05.
The fact that I can designate a specific power set point and have our propulsion system utilize a power near the specified set point tells me that I have met my goal by having designed an active control system that monitors the power consumption of an SAE Aero Competition aircraft.
Unfortunately, although my approach seems to work quit well and despite the fact that the SAE Aero Committee is an advocate of interdisciplinary collaboration and innovative thinking. They have denied me the opportunity of implementing my control system because it would be difficult for them to determine whether my P.I controller complies with the Competition rules.
Recent Comments