Skip to Main Content
This paper presents a hardware-in-the-loop experimental rig for testing a proton exchange membrane fuel cell (PEMFC) driving an electrical vehicle. The PEMFC includes an air-management system with a scroll compressor driven by a high-performance sensorless induction motor drive. The sensorless technique is based on a neural-network-based speed observer, i.e., the total least-squares EXIN full-order observer. The whole system is driven by a classic European Driving Cycle (EEC Directive 90/C81/01). An experimental rig has been built to test the system. The torque-speed characteristics of a real scroll compressor have first been measured. Then, these characteristics have been emulated by a brushless interior-mounted permanent-magnet machine controlled in torque. This emulated scroll compressor has been driven by a sensorless field-oriented controlled induction motor drive. The experimental results show that the system attains a global efficiency of about 50% and that the speed estimation accuracy is high at both very low and very high speeds with a stable behavior at zero speed at no load, which is particularly difficult to achieve for model-based sensorless techniques.