Reliability analysis of a microcontroller board for electric vehicles application using Failure Mode and Effect Analysis (FMEA)

With the increasing presence of automated systems in everyday life, the issue of determining their safety level keeps gaining importance in the design and production cycle. When introducing electrical software driven components into a mechanical system, the safety issue must be addressed both from a structural point of view and from a failure response point of view. The Failure Mode and Effects Analysis (FMEA) is therefore introduced as a crucial step of the production process. The FMEA consists of a detailed study for each component in the system to evaluate its behaviour in case of failure. The output of the analysis comprises a list of failure modes (ways in which a component can fail) ordered from most to least dangerous. The objective of this work was to study the performance of an electric vehicle motor control board in case a failure to any of its components occurs. The first step consisted of evaluating each components failure rate using FIDES to classify all of them in order based on their likelihood to fail. This helped to decrease the analysis complexity since only the components with high failure rates were considered whilst performing FMEA. FMEA was carried out using Risk Priority Number (RPN) evaluation which consists of assigning to all failure modes a level of Severity, Occurrence and Detection, the RPN corresponding to the failure mode is obtained as the result of the multiplication of the mentioned levels. The overall system RPN was then computed as a weighted average of the single components RPN, with weights selected in order to give more relevance to those failure modes that had a high Severity and Occurrence product. The result of the RPN data analysis led to conclude that the system was a high-level risk item. The FMEA was concluded by selecting system components with higher RPN to be improved. Suggestion for changes to the microcontroller board were given considering the cost impact on the system design and the ease with which said changes could be performed. Future research could improve the microcontroller board even further by studying the systems thermal behaviour, with the goal of maximising heat exchange to reduce risk of failure for many of the system components.