Experimental Validation of Fault-Tolerant Control With Signal-Based Diagnosis Approach for Interleaved Boost Converter in Fuel Cell Applications

Abstract

The reliability of dc/dc converters remains a major concern in hydrogen-powered electric cars. As the fuel cell is the main power source, it requires a boost converter to increase its low voltage and adapt it to the dc bus voltage. To ensure continuity of service and fault tolerance of the power converter, it is essential to interleave the phases of the converter's architecture. However, since the semiconductors are fragile, a fault-tolerant control is required to detect and identify the fault, isolate the faulty phase and reconfigure the healthy phases to maintain optimal operation. This is the subject of this article, which is an extension of a previous conference work, on signal-based fault detection and identification method. The proposed fault-tolerant control has been implemented in MicroLabBox 1202 using its processor and internal FPGA board. This approach does not require additional sensors or high sampling rate and is easy to implement. It can even be extended to other multilevel converter topologies.