Parameter Identification based Health Monitoring for a Primary Flight Control Electromechanical Actuator

Our objective is to develop a monitoring algorithm that achieves early detection of degradations that may lead to jamming in the electromechanical actuators (EMAs) used for primary flight surface control. To this end, the typical measurement sequences associated to pre flight test profiles are used to estimate the parameters of a friction model, and hence follow potential friction increase. A sensitivity analysis of the measurement sequences with respect to the friction parameters is conducted and it leads to the segmentation of the measurement sequence in different zones. It is shown that identifying specific parameters in the zones where they exhibit the highest sensitivity allows obtaining estimates with low bias, while identifying the entire model at once using the whole data set leads to highly biased estimates. This illustrates the interest of sensitivity guided estimation for this closed-loop identification problem based on a grey-box EMA model.