Deep learning GAN-based data generation and fault diagnosis in the data center HVAC system

•Generative adversarial network is proposed to generate the training dataset for the building energy system.•The imbalanced dataset issue of FDD model is solved through GAN-based adversarial learning to avoid missing alarms.•An incremental learning SVM using optimal 8 features is developed to diagnose the refrigerant leakage faults in data center HVAC system.•Offline model updating strategy is developed and validated using the experimental datasets.