Abstract: Flywheel energy storage systems have standby losses and are unsuitable for long-term energy storage. Aiming at the economic index of flywheel loss, based on the small sample data of the operation of the flywheel energy storage system, a flywheel loss calculation model combining a Logistic chaotic sparrow optimization algorithm and a convolutional neural network was proposed. Firstly, the causes of flywheel loss were analyzed. Next, the flywheel operation data of Ningxia Lingwu Power Plant was preprocessed, and the adversarial generation network was used for small-sample enrichment. Then, the loss model is established based on the convolutional neural network, the improved sparrow algorithm is used to optimize the hyperparameters of the model, and the superiority of the model is verified by comparison. Finally, simulation experiments show that the model can optimize the output of the flywheel energy storage system and reduce the flywheel loss.