Traditional manual design of Earth Observation Satellite Scheduling Problem algorithms costs massive expert labor. This research introduces AgentAD, an LLM-powered five-agent collaborative framework that converts natural language scenario descriptions into executable Python scheduling codes via four sequential phases. Tests prove its self-evolved algorithms outperform mainstream human-designed heuristics, with memory stream and chain-of-coding mechanisms stabilizing iterative optimization for both basic and complex satellite task scenarios.

Tired of slow, inflexible wind farm maintenance scheduling? New research published in Engineering presents LLM4M, a dedicated fine-tuned language model for turbine repair planning. It quickly generates cost-efficient crew routes with tiny errors versus mathematical optimal solutions, beating popular general LLMs. It runs offline on ordinary hardware and adapts to dynamic real-time maintenance tasks, while its discovered performance phase transition offers practical tips for training industry-specific AI models.

Published in Engineering, this review systematically sorts interdisciplinary engineering therapies for spinal cord injury. It analyzes the three-stage secondary pathological cascade after trauma and elaborates five core intervention branches including neuromodulation, cell transplantation, and functional biomaterials. The article also identifies major translational obstacles and advocates integrated multimodal treatment plus cross-field cooperation to speed clinical transformation of SCI repair technologies.

A newly published review in Engineering defines mitochondrial deacetylase SIRT3 as the core ‘‘head goose molecule’’ triggering negentropic mechanisms to reverse diabetes, obesity, and steatotic liver disease. It sorts three classes of SIRT3 activators and notes translational hurdles, providing systematic theoretical support for targeted metabolic disease therapies.

This report covers a single-cell RNA sequencing study published in *Engineering*, exploring how dietary phytosterol oxidation products (POPs) trigger aortic inflammation in atherosclerosis. The paper identifies 7-ketositosterol (7-KS), the major food-derived POP, drives macrophage M1 polarization via the TLR4–IRF5 signaling axis. Mouse assays and public human arterial single-cell datasets verify this conserved inflammatory molecular signature, offering a single-cell research paradigm for dietary lipid-related cardiovascular disease.

Published in Engineering, this meta-analysis of nine bioreactor datasets uses null models to unpack microbial community assembly rules. It clarifies how stochastic and deterministic processes shift under different environmental disturbances, revealing an inverse link between |βNTI| values and pollutant removal efficiency. The paper builds a three-stage conceptual succession model and offers practical guidance for stable biotreatment operation.

Published in Engineering, this report introduces a full-process intelligent arch-forming technology for large-span mountain arch bridges. Built on three core modules of optimized cable force calculation, digital prefabrication control and automatic installation pose adjustment, the framework replaces traditional physical assembly and repeated manual calibration. Its reliability is verified on the 504 m Deyu Expressway Wujiang Bridge, offering efficient, low-labor construction solutions for prefabricated segmental arch bridges and pointing out follow-up research directions for concrete arch structures.

Published in Engineering, this report introduces an ALK/PEM hybrid electrolyzer multi-timescale scheduling strategy. The study combines CNN–BiLSTM–Attention wind–solar forecasting and hydrogen demand response to refine electrolyzer state modeling via Big-M linearization. Rolling optimization on 1-hour day-ahead and 15-minute intraday timescales stabilizes operation. Tests in Hubei show over 96% schedule execution, lifting system revenue and retaining 80% economic efficiency under extreme cloudy weather.