search
physics-informed AI
Trends
- 1Physics-informed AI tool developed to predict wildfire spread in real time▼A new physics-informed AI tool aims to predict wildfire spread in real time and assist wildfire response
Researchers have unveiled a physics-informed artificial intelligence tool designed to forecast wildfire spread in real time. By combining physical models of fire behaviour with machine learning, the tool aims to give emergency responders faster, more accurate predictions of how blazes will move, helping them plan evacuations and deploy resources more effectively as wildfires intensify worldwide.
- 2
Researchers have developed an AI surrogate model that incorporates built-in physics to accelerate the design of low-power amplifiers. By embedding physical laws directly into the model, the approach reduces the need for slow, computationally heavy simulations, cutting design time significantly. The work highlights a growing trend of physics-informed machine learning being applied to electronics engineering, where it could shorten development cycles for energy-efficient amplifier circuits.
- 3Boosted neural network models computer virus spread▼Boosted physics-informed neural network tracks computer virus spread with vertical transmission
Researchers have developed a boosted physics-informed neural network capable of tracking the spread of computer viruses, including cases with vertical transmission, where malware passes from a system to its derivatives. The approach combines physics-based modeling with machine learning to better predict how infections propagate through networks. Coverage so far is limited to a single specialist outlet, and details on validation or real-world testing are not yet available.
- 4Physics-informed AI reads concrete fracture parameters from stress images▼Physics-informed AI extracts concrete fracture parameters from stress images
A physics-informed artificial intelligence approach can extract fracture parameters of concrete directly from stress images, combining machine learning with physical models of material behaviour. The technique could help engineers assess cracking and structural integrity in concrete faster and more reliably than purely data-driven methods, with applications in construction monitoring and infrastructure safety inspection.
- 5Physics-informed AI cuts CO2 in facade design▼Physics-informed AI reduces operational CO₂ in façade design study
A new study reports that physics-informed AI can reduce operational carbon dioxide emissions in building facade design. By embedding physical laws into the AI model, designers can optimise facades for energy performance and lower emissions during a building's use. The findings highlight growing interest in combining machine learning with building physics to cut the carbon footprint of the construction sector.
- 6Baker Hughes turns to physics-informed AI for compressor simulations▼How Baker Hughes is levering the Dell AI Factory with NVIDIA to Accelerate Centrifugal Compressor Simulations with Physics-Informed AI
Baker Hughes is working with the Dell AI Factory with NVIDIA to speed up simulations of centrifugal compressors using physics-informed AI. The approach combines engineering physics with machine learning to cut the time needed to model compressor performance, a key task in designing equipment for oil and gas operations. The collaboration highlights growing use of AI in industrial engineering workflows.