System and method for controlling fuel cell hybrid electric vehicle with sensor fault tolerance

A fault-tolerant control system for a fuel cell hybrid electric vehicle (FCHEV), comprising:

• a fuel cell, a battery, and an ultracapacitor connected to a direct current-bus (DC-bus) via a plurality of power converters;
• a plurality of sensors configured to measure a current of the fuel cell, a current of the battery, and a current of the ultracapacitor, and a voltage of the DC-bus;
• a controller operatively connected to the plurality of sensors and the plurality of power converters, wherein the controller is configured to:
• calculate a duty cycle value for each power converter in the plurality of power converters using a fractional-order sliding mode control strategy with finite-time stability;
• use a radial basis function neural network (RBFNN) to estimate and compensate for a sensor fault occurring in at least one of the measured currents;
• and
• utilize a minimum learning parameter scheme configured to adjust the RBFNN through adaptation of a single parameter i=ΞiTΞi, (i=1, 2, 3, 4), where ΞiT is a transpose of a weight vector, Ξi is a weight vector, instead of a plurality of neural network weights;
• and
• regulate a current distribution from the fuel cell, the battery, and the ultracapacitor to the DC-bus while maintaining the DC-bus voltage at a reference value despite the sensor fault.