The prediction of gas turbine (GT) future health state plays a strategic role in the current energy sector. However, in the case of limited historical data, e.g., a new installation, training an accurate prognostic model may be challenging. To this aim, this paper develops a Generative Adversarial Network (GAN) model aimed to generate synthetic data that can be used for data augmentation. The GAN model includes two neural networks, i.e., a generator and a discriminator. The generator aims to generate synthetic data that mimic the pattern of real data. The discriminator is a binary classification network. During the training process, the generator is optimized to fool the discriminator in distinguishing between real and synthetic data. The real data employed in this paper were taken from the literature and gathered from three GTs. Data refer to two quantities, i.e., corrected power output and compressor efficiency, which are tracked during several years. Three different analyses are presented to validate the reliability of the synthetic dataset. First, a visual comparison of real and synthetic data is performed. Then, two metrics are employed to quantitively evaluate the similarity between real and synthetic data distributions. Finally, a digital twin (DT) model is trained by only using synthetic data and then it is employed for the prediction of real data. The results prove the high reliability of the synthetic data, which can be thus exploited to train a DT model suitable for prognostic purposes. In fact, the prediction error of the DT model on the real data is lower than 2.5% even in the case of long-term prediction.
Data-driven Generative Model Aimed to Create Synthetic Data for the Long-Term Forecast of Gas Turbine Operation
Losi E.;Manservigi L.;Spina P. R.;Venturini M.
2024
Abstract
The prediction of gas turbine (GT) future health state plays a strategic role in the current energy sector. However, in the case of limited historical data, e.g., a new installation, training an accurate prognostic model may be challenging. To this aim, this paper develops a Generative Adversarial Network (GAN) model aimed to generate synthetic data that can be used for data augmentation. The GAN model includes two neural networks, i.e., a generator and a discriminator. The generator aims to generate synthetic data that mimic the pattern of real data. The discriminator is a binary classification network. During the training process, the generator is optimized to fool the discriminator in distinguishing between real and synthetic data. The real data employed in this paper were taken from the literature and gathered from three GTs. Data refer to two quantities, i.e., corrected power output and compressor efficiency, which are tracked during several years. Three different analyses are presented to validate the reliability of the synthetic dataset. First, a visual comparison of real and synthetic data is performed. Then, two metrics are employed to quantitively evaluate the similarity between real and synthetic data distributions. Finally, a digital twin (DT) model is trained by only using synthetic data and then it is employed for the prediction of real data. The results prove the high reliability of the synthetic data, which can be thus exploited to train a DT model suitable for prognostic purposes. In fact, the prediction error of the DT model on the real data is lower than 2.5% even in the case of long-term prediction.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
