Summary:
UCLA Researchers from the Department of Electrical and Computer Engineering have developed a novel software leveraging advanced machine learning methods to simulate and design high-power laser systems.
Background:
High-power laser systems are crucial to many established industries and in cutting edge research. These systems can be used in manufacturing for metal cutting and additive processes, in materials processing and testing, in medical imaging, and in defense platforms for local delivery of high energy. However, laser systems are expensive to assemble and validate due to the multiple linear and nonlinear optical components they are comprised of, which can limit adoption and development of novel applications. Emerging software solutions have allowed researchers to efficiently design and develop novel laser systems with optimized properties. There is a need to create modeling software that can be used to simulate the complex physical characteristics of high-power laser systems and is compatible with the wide range of optical components available for these systems
Innovation:
Professor Sergio Carbajo and colleagues in UCLA’s Department of Electrical and Computer Engineering have developed a new start-to-end software model which can render electromagnetic fields as well as the energy and spectral distribution of laser systems. Their modular software package features ground-up system design, component selection and configuration, as well as a diagnostics set up. In addition, it also offers physical simulations for each component over wide parameter ranges, generating large amounts of data from the simulated laser. They have shown that this simulated data can be used in machine learning applications for the inverse design of new optical components. Overall, this software has significant potential to streamline laser system engineering for many applications.
Potential Applications:
• Precision welding/manufacturing
• Medical imaging systems
• LIDAR navigation
• Optical component manufacturing
• Defense/military applications
Advantages:
• Start to end system design
• Technology agnostic
• Couples linear and nonlinear optical processes
• Enables machine learning-driven reverse engineering
Development-To-Date:
Researchers have implemented the software and evaluated its performance by simulating the temporal and spectral properties of a photo-injector laser.
Related Papers:
• Hirschman, J., Lemons, R., Wang, M., Kroetz, P., and Carbajo, S. Integrated Software Model for Design, Optimization, and Reverse Engineering of High-Power Laser Systems. CLEO, paper FW3M.2, (2023)
Reference:
UCLA Case No. 2024-067
Lead Inventor:
Professor Sergio Carbajo Garcia, UCLA Department of Electrical and Computer Engineering