Figure 5 | Scientific Reports

Figure 5

From: High precision implicit function learning for forecasting supercapacitor state of health based on Gaussian process regression

Figure 5

The predictions of the GPR models with different mean functions and compound covariance functions, where the first 5% cycles are used for training. The dotted line indicates the 500th cycle. (a,c,e) are the predicted capacitances of an SC for Models 4, 5, and 6, while (b,d,f) denote the predicted errors (observed − mean). The green and blue curves represent the observed and predicted mean capacitance, while the green dots and blue shaded areas denote the predicted training data and confidence interval, respectively. The selection of ± 1.96 for confidence interval is based on Eq. (24).

Back to article page