Accurate fault diagnosis in analog circuits faces significant challenges owing to the inherent complexity of fault data patterns and the limited feature representation capabilities of conventional methodologies. Addressing the limitations of current convolutional neural networks (CNN) in handling heterogeneous fault characteristics, this study presents an efficient channel attention-enhanced multi-input CNN framework (ECA-MI-CNN) with dual-domain feature fusion, demonstrating three key innovations. First, the proposed framework addresses multi-domain feature extraction through parallel CNN branches specifically designed for processing time-domain and frequency-domain features, effectively preserving their distinct characteristic information. Second, the incorporation of an efficient channel attention (ECA) module between convolutional layers enables adaptive feature response recalibration, significantly enhancing discriminative feature learning while maintaining computational efficiency. Third, a hierarchical fusion strategy systematically integrates time-frequency domain features through concatenation and fully connected layer transformations prior to classification. Comprehensive simulation experiments conducted on Butterworth low-pass filters and two-stage quad op-amp dual second-order low-pass filters demonstrate the framework’s superior diagnostic capabilities. Real-world validation on Butterworth low-pass filters further reveals substantial performance advantages over existing methods, establishing an effective solution for complex fault pattern recognition in electronic systems.

To elucidate the long-term changes and driving mechanisms of soil moisture on the Loess Plateau under large-scale ecological restoration, this study analyzed 0–200 cm profile soil water storage from 2000 to 2024 together with precipitation, temperature, and NDVI data. Theil–Sen trend estimation, the Mann–Kendall test, partial correlation analysis, and residual trend analysis were employed to systematically investigate the spatiotemporal variations in soil water storage, its responses to climatic and vegetation factors, and the relative roles of climatic and non-climatic factors. The results showed that soil water storage on the Loess Plateau exhibited pronounced spatial heterogeneity and an overall increasing trend during 2000–2024. Areas with significant and non-significant increases accounted for 52.22% and 41.25% of the study area, respectively, whereas decreasing areas accounted for only 6.53%. Soil water storage was generally positively correlated with precipitation and NDVI, with significant positive correlations covering 80.4% and 90.6% of the study area, respectively. In contrast, soil water storage was predominantly negatively correlated with temperature, with significant negative correlations occurring across 92.3% of the study area. Areas with increasing residual trends accounted for 85% of the study area. The mean relative contributions of climatic and non-climatic factors to changes in soil water storage were 47% and 53%, respectively, indicating that soil water storage dynamics were jointly influenced by hydrothermal climatic conditions and non-climatic factors, including vegetation restoration and land-use change. Overall, soil moisture conditions within the 0–200 cm soil profile of the Loess Plateau improved over the past 25 years, although localized water limited areas remained at risk of soil water storage decline. Ecological restoration should therefore account for regional water resource carrying capacity and optimize vegetation types and planting density to promote coordinated development between ecosystem restoration and water security.