Sensor Fault Diagnosis for BMS of Solar Unmanned Aerial Vehicles With Communication Limitation

ABSTRACT This article investigates a sensor fault diagnosis issue for the battery management system (BMS) of solar unmanned aerial vehicles (SUAV) with communication bandwidth limitation. In practice, the current and voltage sensor faults will lead to an inaccurate estimate of the state of charge and further disturb energy optimal scheduling for the BMS of SUAV. Moreover, a temperature sensor fault may also cause ineffective thermal management and potential safety hazards. To this end, a fault estimation filter based on the sliding mode observer is developed to estimate the current, voltage and temperature sensor faults. A low pass filter is utilized to extract the equivalent output error injection signal generated by the designed sliding mode observer and then achieve detection and estimation of the considered sensor faults. To handle the communication bandwidth limitation for BMS in a desired operating condition, a periodic event‐triggered mechanism is proposed to alleviate communication cost while still retaining the satisfactory fault diagnosis performance. Finally, simulation result demonstrates the effectiveness of the proposed scheme.