The Current electric power grid is undergoing a transition due to the rapid penetration of distributed energy resources (DERs) including renewable energy systems, energy storage systems, and electric vehicles. However, new cybersecurity threats arise. It is still challenging to model and manage a cross-layered security perimeter in multiparty-involved DER systems. Maloperation or malicious control of DERs will be caused by advanced attackers (e.g., hackers and insiders) as seen the real-world attack cases using the expanded attack surfaces. Besides, the quantum era is coming soon and it is anticipated that quantum computing attacks will be possible within 5-10 years. In addition, many inverter-based DERs exposed to the public are vulnerable to physical attacks. To address major threats facing the cyber-physical DERs, this project aims to develop a resilient cyber-physical security framework with a collaborative partnership across multidisciplinary team members from two minority-serving institutions, Texas A&M University-Kingsville and the University of Illinois Chicago, and Sandia National Laboratories. Moreover, this partnership supports an integrative research and education program of the MSIs for training skilled next-generation workforce in the cyber-physical power and energy systems areas.

The technical goal of this project is to launch a major research direction to develop an innovative resilient cyber-physical security framework that addresses imminent challenges in both future cyber-physical security requirements and power engineering designs and resilient operational strategies for DER-rich power systems. Specific integrated research thrusts are as follows: (a) developing a blockchain security governance model for DER systems operating under multiparty and system of systems environments; (b) developing a quantum secure DER network by studying lightweight post-quantum cryptography against quantum computing attacks; (c) realizing DER inverter hardware hardening by investigating a new DER smart inverter security design; and (d) achieving controlled resilience at grid edge using event-triggered resilient self-learning control with retrieval strategy ensuring reduced dependency and susceptibility to communication during the security breach. 

The broader impact/commercial potential of this Partnerships for Innovation - Research Partnerships (PFI-RP) project is to increase dependable power generation from renewable energy resources and to improve the reliability and stability of the nation's electric power grid. Technical problems in connecting and controlling renewable energy resources have resulted in many large-scale failures and the interruption of energy production from these resources, jeopardizing the safe and reliable operation of the nation's electric power system. The proposed technology seeks to overcome the limitations of existing systems, increase reliability for connecting the inverter-based resources to the electric power grid, and allow more renewable energy to supply American homes and businesses. Attaining these goals will increase customer adoption of electric or plug-in hybrid electric vehicles. The technology may increase the competitiveness of US companies in the global inverter, artificial intelligence (AI), and renewable energy markets. The commercial impacts of the project are augmented by an educational outreach plan that will engage student researchers in engineering design problems that integrate and address business-relevant constraints and customer needs.

The proposed project aims to translate AI inverter control innovation into a commercial product, process, and/or service. There are two main gaps to overcome in the proposed project. One is a knowledge gap to extend the previous neural network control technology to practical grid-following and grid-forming applications supporting the operation of inverter-based resources in grid-tied, islanded, and standalone conditions. The other gap is a technical barrier to moving the proposed innovation out of the laboratory into real-world systems. Regarding the knowledge gap, research will equip the neural-network control innovation with modules that can support the inverter operation to meet various grid needs. Regarding the technical barrier, research will be performed in this project to enable the developed neural network inverter to satisfy industry standards. The expected results include prototypes that can be applied to practical electric utility systems and AI-driven inverter technology that can meet the strict reliability requirements of the electric power industry. The commercialization plan consists of a set of guided technology assessments and staged milestones with industry partners on commercial potential and business plan iteration.

Recently deep learning has succeeded mainly in image processing, language processing fields. However, in the real-time control field, deep learning has just started to challenge the dominant role of proportional-integral-derivative controllers in industrial applications, e.g., real-time control of power converters for renewable energy integration. Many urgent problems including training difficulty, the implementation challenges on embedded devices, are curbing deep learning from the development and implementation in embedded control settings. To overcome these difficulties, this project aims to develop novel scalable training algorithms and novel deep neural network controller architectures to fit the strict requirement of embedded control settings.

The interdisciplinary project will develop scalable and reliable deep learning-driven embedded control of power converters in real-time for integrating renewable energy such as solar power. Specifically, this project aims (a) to develop scalable, parallel, fast training algorithms for high sampling frequency, and long-time duration trajectory learning using an high performance computing or cloud platform that will significantly reduce training time from several days, even weeks to several hours, (b) to develop novel deep neural network architectures that can be implemented in embedded devices, e.g., Digital Signal Processors / Field-programmable Gate Arrays without compromising the neural network generalizability and extra computing power and storage requirements.

The project will build and enhance interdisciplinary and inter-institution collaborations between two Minority Serving Institutions: Texas A&M University-Kingsville and North Carolina A&T State University. The project will attract, retain, and educate more minorities particularly Hispanic, African-American, and female students to attend Ph.D. programs. The developed new training algorithm and new architectures for embedded control can be extended to other fields, e.g., bioinformatics, image, robotics, etc. The developed technologies will result in deep learning-driven intelligent control for grid integration of renewable resources and help solve the urgent need to integrate more renewable energy into the power grid in the United States. 

This project is to develop an innovative curriculum on heterogeneous programming for minority students in The Department of Electrical Engineering and Computer Science at Texas A&M University-Kingsville. The new course will cover the introduction of heterogeneous programming and hands-on experiments using intel platforms. The deep learning-based control applications in renewable energy fields from the PI’s previous research will also be incorporated into this new course.