Under the dual imperatives of environmental protection and sustainable socio-economic development, the high energy consumption of the manufacturing sector has become a significant barrier to its transition toward industrial ecological transformation. Integrating machine tool energy-saving technologies into engineering education serves as a pivotal strategy for enhancing energy efficiency and reducing operational consumption while fostering professionals who possess both sustainability awareness and technical expertise. Building on the core principles and future applications of energy-efficient machining, this study introduces a targeted curriculum reform. In response to challenges including cross-disciplinary content, heavy theoretical demands, and insufficient experiential learning, a blended teaching framework is designed, integrating virtual instruction with practical factory-based training. Implementation outcomes demonstrate that this reform significantly enhances students’ comprehension of energy-saving mechanisms and their capacity to apply sustainable concepts in practical engineering contexts. This study provides a reference for integrating energy-saving education into higher education under the framework of sustainable development. It also offers educational support for the manufacturing industry’s pursuit of low-carbon transition and industrial ecological development, demonstrating the important role of engineering education in advancing ecological civilization and contributing to the broader environmental governance system.

Digital transformation and transparency in the seed supply chain are cornerstones of national food security and sustainable agricultural development. Existing agricultural traceability systems suffer from elevated storage overhead and performance degradation with massive seed data processing and fail to iterate quality supervision standards without disrupting continuous business operation. To address these problems, this study proposes a dual-optimization architecture-based traceability system for seed supply chains. An edge-assisted Merkle-tree dimension-reduction aggregation protocol is introduced to compress seed logistics scanning data before blockchain submission. Instead of storing each circulation record as an independent on-chain state update, the proposed scheme anchors one fixed-size Merkle root for each aggregated batch, reducing the per-batch on-chain payload to a constant size and lowering the overall on-chain anchoring burden from record-level growth to batch-level growth. Furthermore, it adopts a decoupled regulatory architecture based on the Strategy Pattern for the separation of traceability state storage and compliance inspection logic, enabling uninterrupted rule switching under the tested upgrade scenario via on-chain hash pointer adjustment. Rigorous statistical evaluation of the experimental results indicates that the system stably processes seed circulation records at a peak effective throughput of 1952.4 transactions per second. Under high-frequency concurrency, the 95th percentile (P95) latency remains controlled under 0.28 s. The average physical on-chain storage for 100,000 circulation records was reduced to 0.52 MB, and deploying a new quality inspection rule takes an average of only 2.2 s, with limited computational resource overhead.