2
$\begingroup$

I have a question regarding the redundancy of hydraulic systems used to deflect the control surfaces of a Boeing 777 aircraft (or the like). We know that "the elevators, ailerons, and flaperons are controlled by two actuators per surface [and] the rudder is controlled by three" [1].

Let's say we have two duplicate hydraulic actuators that control one of the ailerons. How does the Fly-By-Wire (FBW) system ensure that the movement of both actuators are synchronized; in other words, how do the Power Control Units (PCUs) prevent over-extension of one actuator relative to the other so that torsional stresses on the aileron are not exceeded?

I read about LVDT sensors for position feedback being used to control the displacement of the hydraulic pistons with precision. In a hypothetical case we can say that a simple proportional controller would provide a correction to minimize the error, and this error is proportional to the difference between the commanded displacement and the actual displacement of the actuator. Due to differences in resistance between each hydraulic actuator, would a simple proportional control scheme not be sufficient? I envision a more enhanced, "coupled" control scheme to ensure synchronized movement of both actuators:

\begin{equation} PCU_1 = K(p_c - p_1) + G(p_2 - p_1)^n \end{equation} \begin{equation} PCU_2 = K(p_c - p_2) - G(p_2 - p_1)^n \end{equation}

where $PCU_1$ and $PCU_2$ can represent the voltage signal of the solenoid-controlled servovalve for actuators 1 and 2, respectively. $K$ and $G$ are arbitrary proportional gains. $p_c$ denotes the commanded (synchronized) position of both actuators. $p_1$ and $p_2$ are the actual positions of actuators 1 and 2, as measured by their respective LVDT sensor. Lastly, the exponent $n$ can be an arbitrary integer; the higher the value the more the controller will try to compensate any deviations of $p_1$ and $p_2$, so that the difference $p_2 - p_1$ converges to zero more quickly.

Does this coupled logic sound reasonable? Do current commercial jet transport aircrafts employ a similar PCU control scheme for synchronization of multiple actuators connected in duplicate fashion?

*Note that in the above scenario I am assuming each actuator is operating in Active mode, and the servovalves have a closed center configuration to allow for hydraulic lock when no further Actuator Control Electronics (ACE) command is needed. Also note that each actuator has its own independent hydraulic circuit, pump, reservoir, valve, sensor, wiring, etc. In addition, I'm assuming the actuators are separated along the aileron's span. Each control rod deflects the aileron at a common hinge point and equivalent moment arm, hence why I chose $p_c$ to be the same for both actuators. (The commanded positions of actuators 1 and 2 could still be independent.)

[1]: Boeing B-777: Fly-By-Wire Flight Controls

$\endgroup$
7
  • $\begingroup$ It would make sense to let each actuator have it's own position sensor simply to make the wiring simpler. $\endgroup$ Jul 31, 2023 at 0:35
  • $\begingroup$ The point of multiple independently-controlled actuators is to provide voting between multiple control systems without a single point of failure. The fact that you have a single command, $p_c$, defeats this. To use this technique, you simply make the structure of the aileron strong enough to withstand conflicting commands. Some systems use self-monitoring: if the surface isn't moving to the position the loop wants, then the loop realizes it's being out-voted and it trips itself offline. $\endgroup$
    – user71659
    Jul 31, 2023 at 0:38
  • $\begingroup$ @user3528438 sorry for not making this clear. From the scenario I'm assuming each actuator has its own unique hydraulic circuit, valve, sensor, wiring, etc. $\endgroup$ Jul 31, 2023 at 0:50
  • 1
    $\begingroup$ Check out section 11.8.5 in the document you linked; that seems to discuss the scenario you're asking about. $\endgroup$
    – Ralph J
    Jul 31, 2023 at 19:37
  • 1
    $\begingroup$ @RalphJ WOW... I can't believe the answer I was looking for was right under my nose! Thank you for pointing out that section, which is something I overlooked. It makes a lot more sense to monitor the difference in fluid pressure of the two actuators--as opposed to the difference in their positions--to avoid overstressing the individual control surfaces due to unequal actuation forces. $\endgroup$ Aug 1, 2023 at 19:56

0

You must log in to answer this question.