Design of an Integrated System for On-line Test and Diagnosis of Rotary Actuators
- PDF / 1,588,254 Bytes
- 7 Pages / 595.224 x 790.955 pts Page_size
- 28 Downloads / 169 Views
Design of an Integrated System for On-line Test and Diagnosis of Rotary Actuators Leonardo B. Piccoli1
· Renato V. B. Henriques2 · Tiago R. Balen1
Received: 16 May 2019 / Accepted: 1 July 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract In this paper, the design of an on-chip Fault Detection and Diagnosis System for Condition Based Maintenance of electromechanical actuators is presented. The proposed system is based on signal processing algorithms integrated in a customized Application Specific Integrated Circuit (ASIC). The design was synthesized using a 90nm CMOS standard cell library. As a case study, post-synthesis simulations were performed using signals acquired from a real electromechanical valve, using torque and vibration sensors considering both fault-free and defective situations for the actuator. Results show the effectiveness of the system in performing real-time fault detection and identification, with low power consumption and low silicon area utilization. Keywords Condition-based maintenance · Fault detection · Fault prediction · Diagnosis · Electric valves · On-line test · Electromechanical actuators
1 Introduction On-line fault detection, prediction and diagnosis in mechanical systems is of significant importance, as, for instance, in pipelines for oil or gas transportation and other related industry activities [7, 8]. When considering oil or gas transportation, electromechanical valves are widely used in complex pipelines. Therefore, due to the safety and financial-critical nature of such activities, the fault-free operation of electric valves is crucial. This demands strategies for test and predictive maintenance, which, in electromecanical systems, requires autonomous and continuous
Responsible Editor: K. K. Saluja Leonardo B. Piccoli
[email protected] Renato V. B. Henriques [email protected] Tiago R. Balen [email protected] 1
Graduate Program on Microelectronics, Federal University of Rio Grande do Sul, Porto Alegre, RS, Brazil
2
Graduate Program on Electrical Engineering, Federal University of Rio Grande do Sul, Porto Alegre, RS, Brazil
monitoring of signals acquired in mechanical structures or electric parts of the monitored machines. Several techniques for health monitoring of mechanical equipment have been proposed in the last years. They are usually based on signal processing (using, for example, Wavelet and Fourier transforms and adaptive filtering), artificial intelligence networks and machine learning, as for example, [6, 9, 13]. Such techniques are implemented in personal computers (PC), resulting in large footprints and power-hungry systems. Some other techniques are implemented in Field Programmable Gate Arrays (FPGAs), as for example, [3], in which a fault diagnosis for steam turbine systems, was proposed. In [2], an on-board real-time system to detect and isolate critical faults in an electromechanical actuator of aircraft is proposed. A real-time FPGA-based fault monitoring system of power electronics interfaces in wind e
Data Loading...
