@Article{cmc.2025.068351,
AUTHOR = {Shige Wang, Yalong Liang, Lian Huang, Pei Li},
TITLE = {Cuckoo Search-Deep Neural Network Hybrid Model for Uncertainty Quantification and Optimization of Dielectric Energy Storage in Na<sub>1/2</sub>Bi<sub>1/2</sub>TiO<sub>3</sub>-Based Ceramic Capacitors},
JOURNAL = {Computers, Materials \& Continua},
VOLUME = {85},
YEAR = {2025},
NUMBER = {2},
PAGES = {2729--2748},
URL = {http://www.techscience.com/cmc/v85n2/63842},
ISSN = {1546-2226},
ABSTRACT = {This study introduces a hybrid Cuckoo Search-Deep Neural Network (CS-DNN) model for uncertainty quantification and composition optimization of Na<sub>1/2</sub>Bi<sub>1/2</sub>TiO<sub>3</sub> (NBT)-based dielectric energy storage ceramics. Addressing the limitations of traditional ferroelectric materials—such as hysteresis loss and low breakdown strength under high electric fields—we fabricate (1 − <i>x</i>)NBBT8-<i>x</i>BMT solid solutions via chemical modification and systematically investigate their temperature stability and composition-dependent energy storage performance through XRD, SEM, and electrical characterization. The key innovation lies in integrating the CS metaheuristic algorithm with a DNN, overcoming local minima in training and establishing a robust composition-property prediction framework. Our model accurately predicts room-temperature dielectric constant (<i>ε</i><sub><i>r</i></sub>), maximum dielectric constant (<i>ε</i><sub><i>max</i></sub>), dielectric loss (<i>tan δ</i>), discharge energy density (<i>W</i><sub><i>rec</i></sub>), and charge-discharge efficiency (<i>η</i>) from compositional inputs. A Monte Carlo-based uncertainty quantification framework, combined with the 3σ statistical criterion, demonstrates that CS-DNN outperforms conventional DNN models in three critical aspects: Higher prediction accuracy (R<sup>2</sup> = 0.9717 vs. 0.9382 for <i>ε</i><sub><i>max</i></sub>); Tighter error distribution, satisfying the 99.7% confidence interval under the 3σ principle; Enhanced robustness, maintaining stable predictions across a 25% composition span in generalization tests. While the model’s generalization is constrained by both the limited experimental dataset (n = 45) and the underlying assumptions of MC-based data augmentation, the CS-DNN framework establishes a machine learning-guided paradigm for accelerated discovery of high-temperature dielectric capacitors through its unique capability in quantifying composition-level energy storage uncertainties.},
DOI = {10.32604/cmc.2025.068351}
}