&nbsp; Ubiquitous application, especially in places with harsh environments and strict safety measures, makes Fault Detection and Diagnosis (FDD) in hydraulic actuators an imperative concern for the industry. The advent of new technologies—in the fields like sensors, signal processing, control design, and real-time computation —has opened new research opportunities to replace traditional maintenance methods with automated approaches to manage faults. These approaches can be categorized into two main groups: Model-based FDD and signal-based FDD. Due to numerous studies on the modeling of fluid power systems, model-based FDD is commonly used by researchers.&nbsp;&nbsp;&nbsp;&nbsp; Model-based FDD uses estimation strategies, including observers and parameter estimation tools. The observer-based FDDs generally compare the simulated and actual response of a system and then use the residual to detect and diagnose faults. The main problem here is generating residuals that are specific to fault conditions and could enable fault diagnosis. Parameter estimation can add to residual analysis and provide further information for fault diagnosis. However, observability is a limiting factor in most applications—including fluid power systems. To address the observability problem, adaptive strategies like Interactive Multiple Model (IMM) estimation have been proven to be effective. The concept is to use different models—each one represents a different fault condition—then, to correct the residual from these models to determine a probability factor for fault condition leading to their diagnosis.&nbsp;&nbsp;&nbsp;&nbsp; In this paper a computationally efficient form of IMM referred to as the Updated IMM (UIMM) is used and applied to an Electro-Hydrostatic Actuator (EHA) for FDD. The UIMM is suited to fault conditions that are irreversible, meaning that if a fault happens it will persist in the system. In essence, the UIMM follows through a progression of models that in line with the progression of the fault condition in lieu of having all models being considered at the same time as is the case for IMM. Hence, UIMM reduces the number of models running at the same time significantly, which has two major advantages that are higher computational efficiency and avoiding the combinatorial explosion.&nbsp; The state and parameters estimation strategy that is used in conjunction with UIMM is the Variable Boundary Layer Smooth Variable Structure Filter (VBL-SVSF). The VBL SVSF is a robust optimal estimation strategy that is more stable than the Kalman Filter in relation to system and modeling uncertainties.&nbsp;&nbsp;&nbsp; The EHA system, used here, is a closed-loop circuit that controls a linear hydraulic actuator by adjusting the flow rate by changing the speed of a fixed displacement gear pump. This system is more efficient in both size and energy consumption than the conventional servo-valve controlled hydraulic systems. the focus of this study is on detecting and quantifying leakage and friction faults. The UIMM method is validated by simulation of the fault conditions on an EHA using MATLAB Simulink.

A Model-Based FDD Approach for an EHA Using Updated Interactive Multiple Model SVSF

Paper Proceedings: https://asmedigitalcollection.asme.org/FPST/FPMC2021/volume/85239

2021 ASME/BATH Symposium on Fluid Power and Motion Control (FPMC2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.