Tyan-WP: New Foundation Model Achieves Accurate Wind Power Forecasting Without Site-Specific Training
Researchers have introduced Tyan-WP, a wind power foundation model designed for ultra-short-term probabilistic forecasting that can operate without site-specific training data. The model was pretrained on over 126,000 U.S. wind sites across seven years and incorporates terrain, coordinate, and meteorological metadata to improve accuracy. It addresses a critical industry gap as wind farms proliferate across diverse geographies and require faster grid connection without lengthy data collection periods.
Tyan-WP is presented as the first foundation model specifically designed for wind power ultra-short-term probabilistic forecasting, submitted to arXiv in June 2026. The model tackles two known shortcomings in existing approaches: site-specific time series models struggle in data-scarce scenarios, while generic large time series models are largely limited to univariate inputs and fail to exploit site metadata or power-meteorological dependencies. Tyan-WP addresses these gaps through two domain-specific modules: a static site embedding layer using coordinate, terrain, and ecoregion data, and a power-aware meteorological fusion (PAMF) module. Pretrained on a large-scale U.S. dataset, the model achieves zero-shot forecasting and outperforms eight supervised site-specific models and eleven generic large time series models in benchmarks, reducing MAE by 19.9%, RMSE by 16.6%, CRPS by 22.2%, and AQL by 21.7% while raising R² by 16.7%. The model also demonstrates cross-geography generalization on six real U.K. wind sites, suggesting practical applicability beyond its training distribution.
What's missing
The paper is a preprint and has not yet undergone peer review. Key open questions include: whether the U.S. pretraining dataset is publicly available or proprietary; computational requirements for deployment at scale; and how performance degrades under extreme or rare meteorological events not well-represented in training data.
What different sources said
- arXiv cs.AICenter
Tyan-WP: A Wind Power Foundation Model for Ultra-Short-Term Probabilistic Forecasting
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.