Topology optimization serves as a critical method for promoting lightweight structural design. Traditional methods predominantly focus on mechanical performance evaluation, often neglecting the critical correlation between modal characteristics and structural stiffness. The Evolutionary Structural Optimization (ESO) method is extensively employed in topology optimization; however, iterative oscillations lead to issues such as grid divergence and diminished solution quality. To address issues such as iterative oscillations and mesh divergence in the traditional Evolutionary Structural Optimization (ESO) method, this study applies a Simp Evolutionary Structural Optimization (SI-ESO) methodology. This method integrates intermediate density parameters and penalty factors into the progressive structural optimization process, thereby significantly enhancing iterative convergence and model quality. This work applied the optimized SI-ESO method to the lightweight redesign of an aviation generator’s rear cover, with validation conducted through additive manufacturing. Subsequently, the back cover of an aviation generator was redesigned and fabricated utilizing additive manufacturing technology. Empirical results indicate that under maximum stress conditions and employing the same additive process, the maximum deformation of the SI-ESO-optimized model is reduced compared to that of the ESO-designed model. Compared with the original design, the SI-ESO-optimized model achieved a 31% weight reduction, while relative to the ESO-optimized model, it exhibited a 27% lower maximum stress and a 10.53% higher first-order frequency, demonstrating both lightweighting and enhanced structural stiffness.

To address insufficient detail representation, inadequate cross-modal fusion, and limited localization accuracy in object detection under low-light and complex background conditions, this study proposes an improved CenterNet-based multimodal object detection method. The model uses fused images and infrared images as dual-source inputs, where infrared wavelet priors are introduced to enhance texture and structural representation. A Feature Fusion Attention (FFA) module is designed to improve cross-modal feature interaction, while a Heatmap-Guided Detection Head (HGDH) is introduced to explicitly enhance target-related regions during detection. In addition, a two-stack Hourglass backbone is adopted for multi-scale modeling of global semantics and local details. Based on the public LLVIP dataset, a precisely paired fused-image/infrared-image dataset, termed RH-25, was constructed for experiments. On the RH-25 dataset, the proposed method achieves a 3.51% improvement in mAP@0.5 relative to the baseline, as demonstrated by comparative and ablation experiments. Moreover, supplementary experiments on the MFAD dataset indicate potential cross-dataset adaptability under different scenes and multi-class conditions. These results indicate that the proposed method can improve detection performance in low-light and complex environments.