Artificial intelligence (AI) techniques have been spreading in most scientific areas and have become a heated focus in photonics research in recent years. Forward modeling and inverse design using AI can achieve high efficiency and accuracy for photonics components. With AI-assisted electronic circuit design for photonics components, more advanced photonics applications have emerged. Photonics benefit a great deal from AI, and AI, in turn, benefits from photonics by carrying out AI algorithms, such as complicated deep neural networks using photonics components that use photons rather than electrons. Beyond the photonics domain, other related research areas or topics governed by Maxwell’s equations share remarkable similarities in using the help of AI. The studies in computational electromagnetics, the design of microwave devices, as well as their various applications greatly benefit from AI. This article reviews leveraging AI in photonics modeling, simulation, and inverse design; leveraging photonics computing for implementing AI algorithms; and leveraging AI beyond photonics topics, such as microwaves and quantum-related topics.

In this work, a full-wave partial element equivalent circuit (PEEC) model for 2-D periodic metallic structures is proposed. The mixed potential integral equations (MPIEs) with periodic Green’s function (GF) are interpreted as loop equations following Kirchhoff’s voltage law (KVL) in a circuit domain. The impact of gauge freedom in the MPIE on the corresponding PEEC model is discussed. A tailored gauge function is introduced to form a generalized PEEC model for periodic structures, which demonstrates superior passivity and frequency stability. With the proposed PEEC model, lumped and active elements can be easily integrated with conducting bodies for field-circuit co-simulation. Through the distributed circuit including coupling elements, the PEEC model provides a clear physical insight into the electromagnetic behavior of the physical models for design and optimization. Three numerical examples, including frequency-selective surface (FSS), array antenna, and pixelwise meta-atom, are given to illustrate the accuracy and efficiency of the proposed PEEC model.

In this article, a compact partial element equivalent circuit (c-PEEC) model is presented for characterizing the electromagnetic (EM) problems with finite-sized piecewise homogeneous dielectrics. Unlike in conventional PEEC models where dielectrics are described by massive subcircuits, the proposed c-PEEC model addresses the composite conductor-dielectric problems via circuit elements resident on conductors only. Novel null-field boundary integral equations (n -BIEs) are formulated based on the surface equivalent principle and quasi-static assumption. The elimination of the magnetic field-related equivalent sources avoids the low-frequency breakdown problem caused by the weak EM coupling at low frequencies and improves the accuracy by inhibiting the numerical errors in discretizing magnetic sources. The computational cost is significantly reduced compared with the conventional models. Furthermore, the concise configuration of the c-PEEC model extends the model-order reduction (MOR) algorithms for fundamental homogeneous models to heterogeneous integration. Three numerical examples including a resonator, a radio frequency (RF) embedded passive circuit, and an interconnection problem, are studied to validate the stability, efficiency, and accuracy of the c-PEEC model. A micro-modeling circuit is obtained from the c-PEEC model to demonstrate its compatibility with the MOR algorithms.