Збирач потоків

Infineon Technologies AG of Munich, Germany has expanded its CoolSiC MOSFETs 400V G2 portfolio with the top-side-cooled (TSC) TOLT package as well as the TO-247-3 and TO-247-4 packages. In addition, three new products in the TOLL package have been introduced, with rated voltages of 440V (continuous) and 455V (transient)...

The industrial scenery is getting reshaped by digital twins and physical AI. These virtual replicas of factories, facilities, or even processes were once mainly conceived for planning purposes and now have become more operationally oriented, mainly concerned with training autonomous robots, AI-powered machinery, and operational systems to perform their tasks safely and efficiently in the real world. High-tech OpenUSD, immersive simulation tools, and AI-driven modeling are helping developers create high-fidelity digital twins at scale, removing most of their manual labor and fast-tracking industrial AI deployment.

Scaling Industrial AI and Physical AI with Digital Twins

Digital twins provide a virtual environment within which physical AI agents such as autonomous robots or smart factory systems can learn and adapt before deployment. Simulations of a finer quality came at the cost of much manual effort. Today, with advanced OpenUSD, neural reconstruction, and world foundation models (WFMs), developers can now set about constructing these complex digital replicas far more rapidly.

Key developments include:

SDKs bridging between simulators: They allow people to simulate robots and systems in diverse simulators, thus virtually providing access for robotics developers anywhere in the world.

• Neural rendering and 3D reconstruction libraries: These allow the capture and reconstruction of sensor data from the real world, simulation, and photorealistic rendering.
• Open-source robotics frameworks: Offer readymade environments and schemas for robots and sensors to help reduce the simulator-to-reality gap.
• World foundation models (WFMs): Used to create synthetic datasets and to carry out higher-order reasoning on these datasets for the benefit of physical AI applications.
• Advanced rendering and AI-assisted material modeling: Provide scalable ways to create industrial-grade digital twins.

OpenUSD: Powering the Future of Industrial 3D Innovation

OpenUSD constitutes the backbone of industrial 3D workflows, having become a standard for digital twin creation with interoperability between industrial and 3D data. By now, the Alliance for OpenUSD (AOUSD) has been extended to include Accenture, Esri, HCLTech, PTC, Renault, and Tech Soft 3D, thus showing great endorsement of OpenUSD and present objectives of uniting industrial 3D workflow.

To support this growing ecosystem, NVIDIA has introduced an industry-recognized OpenUSD development certification and a digital-twins learning path, helping developers gain the skills needed to build the factories and industrial systems of tomorrow.

Industry Applications Driving the Future:

Some of the global leaders use digital twins and OpenUSD for transforming industrial operations:

• Siemens: Teamcenter Digital Reality Viewer allows working with large-scale digital twins for visualization and collaboration, thereby reducing physical prototyping and faster time-to-market.
• Sight Machine: Operator Agent platform amalgamates live production data with AI-driven recommendations and digital twins for better plant visibility and faster decision-making.
• Rockwell Automation: Emulate3D Factory Test creates physics-based digital twins from simulation to optimize automation and autonomous systems.
• EDAG: Uses digital twin for project management, production layout optimization, worker training, and data-driven quality assurance.
• Amazon Devices & Services: Uses digital twin environments to train robot arms for assembly, testing, packaging, and auditing, all with no physical intervention.
• Vention: Offers plug-and-play digital twin and automation solutions so intelligent manufacturing systems can be deployed more speedily.

Conclusion:

The combination of OpenUSD, digital twins, and AI-driven simulation is transforming industrial operations on the ground. By proving the exact, scalable virtual environment, they allow manufacturers, robot developers, and physical AI engineers to innovate faster, cut down expenses, and systematize safer and smarter solutions faster than ever before.

(This article has been adapted and modified from content on NVIDIA.)

The post OpenUSD and Digital Twins: Transforming Industrial AI Workflows appeared first on ELE Times.

Wolfspeed Inc of Durham, NC, USA — which makes silicon carbide (SiC) materials and power semiconductor devices — has completed its financial restructuring process and emergence from Chapter 11 protection...

Epirus Inc of Torrance, CA, USA says that, in an invitation-only demonstration on 26 August at Camp Atterbury, Indiana (attended by representatives from throughout the US Department of Defense, other US Government agencies and nine allied countries), its Leonidas solid-state high-power microwave (HPM) system — which uses gallium nitride (GaN) transistors to minimize size, weight and deployment time — delivered weaponized electromagnetic interference to counter swarms of robotic, asymmetric threats...

The global energy sector is at a historic turning point. Renewable energy integration, EV promotion, and AI-driven consumption create more demand on already complex grids. The transformation calls for a new era of energy professionals who can build a bridge between traditional engineering and digital technologies-the infrastructure upgrades alone cannot solve the equation.

The Digital Shift in Energy Systems

Modern power systems evolve into interconnected, intelligent networks. Smart grids, real-time balancing, and consumer-driven energy management are redefining how electricity flows. Still, the digital revolution carries many challenges requiring upskilling and interdisciplinary knowledge to solve.

Top Challenges Facing the Next Generation Workforce:

1. Dual-Skill Gap

Engineers today need expertise in network-relevant issues and traditional grid operations, plus in cybersecurity matters. Still, there are few professionals with an engineering background and digital expertise; this scarcity leads to inefficiency in troubleshooting and system reliability.

2. A Shift Toward Virtualization

Careful changes from hardware-based to software-driven operations have increasingly taken protection and control functions onto a virtual platform. Hence, engineers will have to embrace digital tools with data analytics and server technologies that are not traditional to the power area.

3. Cross-system Collaborations

Data exchanges must be smooth as renewable assets such as solar and battery storage interfacing with distribution and transmission networks. Therefore, engineers must manage various protocols and formats, settling voltage, frequency, and power flows after the interface in real time.

Building the Workforce of Tomorrow

Such challenges require: Full-training in digital communication, grid standards such as IEC 61850, and advanced networking.

Simplified Tools and Platforms that reduce technical complexity and enable engineers to focus on system optimization.

Collaborative Ecosystems where power engineers, IT experts, and operators work together to maintain resilience across distributed networks.

Conclusion:

The future of energy will be shaped as much by people as by technology. Companies that invest in digital skills, upskilling programs, and collaborative frameworks will lead the transition to resilient, intelligent grids. Industry leaders such as Moxa, with their training initiatives and global expertise, are playing a vital role in equipping professionals to thrive in this new era ensuring the workforce is ready to power the grids of tomorrow.

(This article has been adapted and modified from content on Moxa.)

The post Future-Proofing the Energy Workforce in a Digitally Driven Era appeared first on ELE Times.

ANRITSU CORPORATION has developed and launched its new 60 GHz optical sampling oscilloscope MP2110A-080 option for the BERTWave MP2110A. This option verifies the performance of 200G/Lane optical transceivers forming the foundation of faster data-center communications and growing AI deployment. It delivers high PAM4 TDECQ evaluation accuracy and measurement productivity for next-generation high-speed optical transceivers, such as 1.6T and 800G, supporting strong quality assurance of large-capacity, high-speed communications infrastructure.

This test solution was exhibited as a reference at the China International Optoelectronic Exposition (CIOE 2025) on September 10, 2025, and will also be showcased at the European Conference on Optical Communication (ECOC 2025), one of the world’s leading international conferences in the field of optical communications, to be held in Copenhagen, Denmark, from September 29 to October 1, 2025.

Development Background
With the growth of AI data centers, optical communication speeds are increasing from 800G to 1.6T, and transmission rates are shifting from 50 Gbaud (100G/Lane) to 100 Gbaud (200G/Lane). As transmission speeds increase, there is a growing need for wideband sampling oscilloscopes capable of evaluating higher frequency components in optical transceiver signals.

Product Features
The all-in-one MP2110A solution integrates the necessary functions for physical-layer evaluation of optical transceivers during development and manufacturing. This new 60 GHz oscilloscope MP2110A-080 option enables evaluation and analysis of next-generation high-speed 200G/Lane communication standards.

• High-Accuracy PAM4 TDECQ Measurement: With the performance of a reference receiver supporting PAM4 signals up to 120 Gbaud, the MP2110A offers reliable TDECQ evaluations by leveraging the high measurement accuracy of existing models.
• Improved Efficiency with Simultaneous 4-Channel Measurement: By measuring four optical signals simultaneously, the MP2110A cuts measurement time and improves operation efficiency. Batch evaluation of multiple channels simplifies measurement systems and processes to enhance productivity.
• Further Productivity Gains with Faster Measurement: Increasing the MP2110A sampling speed fourfold compared to previous models shortens measurement times even further. Stable operation with a built-in PC improves R&D and manufacturing efficiency.
• Cost-Effective 4-Channel Software Upgrade Option: With a software upgrade path to 4-channels, the 2-channel option lowers initial costs, allowing flexible deployment supporting future expansion matching budget and evaluation environment.

The post Anritsu introduces a 60 GHz Optical Sampling Oscilloscope for 200G/Lane 1.6T Transmission appeared first on ELE Times.