Identification of Precursors of Servovalves Failures for Implementation of an Effective Prognostics

Servovalves are critical components of the hydraulic servos and their correct operation is mandatory to ensure the proper functioning of the controlled servosystem. A continuous monitor is typically performed to detect a servovalve loss of operation, but this monitor falls short of recognizing other malfunctionings. Often, a progressive degradation of a servovalve occurs, which does not initially create an unacceptable behaviour, but eventually leads to a condition in which the servovalve, and hence the whole servoactuator operation, is impaired. Developing a prognostic algorithm able to identity the precursors of a servovalve failure and its degradation pattern is thus beneficial for anticipating the incoming failure and alerting the maintenance crew such to properly schedule the servovalve replacement. This avoids a servovalve failure in service, thereby ensuring improved equipment availability and minimizing the impacts onto the logistic line. To this effect, a model based prognostic technique was developed that fuses several informations obtained by comparing actual with expected responses of the servovalve to recognize a degradation and estimate the remaining useful life. The research presented in the paper was focused to develop a technique able to identify symptoms alerting that a servovalve is degrading and will eventually exhibit an anomalous behaviour. To assess the robustness of the technique, an appropriate simulation test environment was developed. Random combinations of time histories of servoactuator commands, loads, continuously varying servovalve offsets, supply and return pressures, hydraulic fluid temperatures were simulated in association with noise acting on the equipment signals and loads. Simulations were then run in which servovalve degradations progressing with an irregular pattern were injected into the servovalve of a flight control actuator while being subjected to the previously defined flight test environment and the ability of the algorithms to correctly sort out the failure precursors was assessed. The results showed an adequate robustness and confidence was gained in the ability to early identify the servovalve malfunctionings with low risk of false alarms or missed failures.