Fault tolerant schemes of non-isolated DC-DC converters for renewable energy applications

The increasing number of modern applications such as electric vehicles, renewable energy systems and portable electronic gadgets, DC-DC converters have attracted much attention as the reliable and efficient power electronic devices. These applications generally use non isolated DC-DC converters due to the cost effectiveness, efficiency and simplicity. But they are susceptible to faults such as power switch failure, capacitor, inductor, decrease in the system reliability and performance. In this review an overall analysis of fault tolerant approaches to enhance the stability of non-isolated DC-DC converters is presented. It analyses several fault detection and diagnosis methods, including as model-based, signal-based and data-driven methods, and fault-tolerant control methods, such as redundant topologies, adaptive modulation and reconfiguration methods. This work analyses the effectiveness, implementation challenges and trade-offs of various methods in terms of cost, complexity and efficiency. By comparing the different strategies it can be shown that a cost-efficient fault tolerant strategy may be achieved without the need of extra hardware implementation techniques. Hardware-based approaches provide faster fault isolation and improved performance after a fault, but are more expensive since they need extra hardware. Furthermore, this review article identifies the current research gaps, such as the lack of post-fault efficiency evaluation and large-scale practical implementations. New domains such as artificial intelligence, machine learning and intelligent fault diagnostic system are also seen as possible future research topics. The present work might be a useful reference for researchers and engineers who are interested in increasing the reliability of fault tolerant non-isolated DC-DC converters in current applications.