Lectures
Tutorials
Applications
“Inverse design” is a technique where one can use computational methods to automate the process of designing photonic devices. The designer simply expresses the desired performance as a mathematical optimization problem and lets the computer find the solution. In this course, we dive into the fundamentals of inverse design, how it is enabled using the “adjoint method”, and give practical examples and important considerations when setting up an inverse design problem. The course includes video lectures and tutorials using Flexcompute's Tidy3D solver for its examples, and the scripts used in the slides are publicly accessible from this website. Basic knowledge of electromagnetics and programming is recommended to get the most use out of this course. A basic knowledge of FDTD (e.g., our FDTD 101 series) is helpful but not required.
After completing this course, you will be able to:
Understand the fundamentals of inverse design.
Have a basic understanding of how gradients are computed efficiently using the "adjoint method".
Get hands-on experience with inverse design optimization examples.
Learn how to include fabrication constraint considerations in your optimization.
Learn intuition regarding how to improve performance and properly set up inverse design problems.
Lectures
In this lecture, we give a broad introduction to the concept of computational design in photonics. We show how simulation and mathematical optimization can be combined to analyze and design photonic devices effectively.
In this lecture, we derive the adjoint variable method for computing the gradient of the figure of merit of a photonic device. We derive this gradient from first principles, starting from the solution of Maxwell’s equations in the frequency domain. Using a simple example of focusing electromagnetic field intensity at a single position, we derive the relevant terms in the gradient and give physical interpretation. We discuss how this method provides gradient information with only two simulations, regardless of the number of design parameters, and how this enables inverse design optimization.
In this lecture, we show how to use the previously introduced “adjoint variable method” to perform gradient-based optimization of a focusing lens. We set up a simple device based on a pixellated array of permittivity values, compute the gradient of this array with respect to the focusing strength of our lens, and then perform an optimization to achieve a final device that focuses light successfully.
In this lecture, we will discuss the need for fabrication constraints in inverse design optimization. We describe one approach to this and demonstrate it by optimizing a Silicon photonics mode converter.
In this lecture, we introduce a method to optimize your device with respect to a geometric parameterization using inverse design and the adjoint method. As an example, we demonstrate the inverse design of a 90 degree waveguide bend by shifting the boundaries of the device.
In this lecture, we introduce a method to optimize your device using a level set parameterization using inverse design and the adjoint method. As an example, we demonstrate the inverse design of a waveguide splitter.
This website uses cookies to ensure you get the best experience. Learn more about our cookie policy.
AcceptThanks for subscribing
TERMS & CONDITIONS
EFFECTIVE DATE: January 1, 2025
Terms and Conditions for User-Submitted Content
By submitting content to Flexcompute, you agree to the following terms:
1. Ownership and Copyright
2. Responsibility for Content
3. Modification and Presentation
4. Liability Disclaimer
5. Content Use Rights
6. Privacy and Confidentiality
Enter your email address below to receive the presentation slides. In the future, we’ll share you very few emails when we have new tutorial release, development updates, valuable toolkits and technical guidance . You can unsubscribe at any time by clicking the link at the bottom of every email. We’ll never share your information.