Detecting cyberattacks in photovoltaic (PV) monitoring and MPPT control signals requires models that are robust to bias, drift, and transient spikes, yet lightweight enough for resource-constrained edge controllers. While deep learning outperforms traditional physics-based diagnostics and handcrafted features, standard fine-tuning is computationally prohibitive for edge devices. Furthermore, existing Parameter-Efficient Fine-Tuning (PEFT) methods typically apply uniform adaptation or rely on expensive architectural searches, lacking the flexibility to adhere to strict hardware budgets. To bridge this gap, we propose Constraint-Driven Warm-Freeze (CDWF), a budget-aware adaptation framework. CDWF leverages a brief warm-start phase to quantify gradient-based block importance, then solves a constrained optimization problem to dynamically allocate full trainability to high-impact blocks while efficiently adapting the remaining blocks via Low-Rank Adaptation (LoRA). We evaluate CDWF on standard vision benchmarks (CIFAR-10/100) and a novel PV cyberattack dataset, transferring from bias pretraining to drift and spike detection. The experiments demonstrate that CDWF retains 90 to 99% of full fine-tuning performance while reducing trainable parameters by up to 120x. These results establish CDWF as an effective, importance-guided solution for reliable transfer learning under tight edge constraints.

翻译：摘要在光伏（PV）监测和最大功率点跟踪（MPPT）控制信号中检测网络攻击，需要模型对偏差、漂移和瞬态尖峰具有鲁棒性，同时轻量级以适应资源受限的边缘控制器。尽管深度学习优于传统的基于物理的诊断和手工特征提取方法，但标准微调对边缘设备而言计算成本过高。此外，现有参数高效微调（PEFT）方法通常采用统一适配或依赖昂贵的架构搜索，缺乏在严格硬件预算约束下的灵活性。为弥补这一不足，我们提出了约束驱动热启动冻结（CDWF）——一种预算感知的适配框架。CDWF通过简短的热启动阶段量化基于梯度的块级重要性，随后求解约束优化问题：动态分配完全可训练性给高影响力模块，同时对剩余模块通过低秩适配（LoRA）进行高效调整。我们在标准视觉基准（CIFAR-10/100）和新型PV网络攻击数据集上评估CDWF，从偏差预训练迁移至漂移和尖峰检测。实验结果表明，CDWF在保留全微调性能90%至99%的同时，可训练参数最多减少120倍。这些结果证明CDWF是一种在严格边缘约束下实现可靠迁移学习的有效、重要性引导解决方案。