Abstract:Synergies between advanced communications, computing and artificial intelligence are unraveling new directions of coordinated operation and resiliency in microgrids. On one hand, coordination among sources is facilitated by distributed, privacy-minded processing at multiple locations, whereas on the other hand, it also creates exogenous data arrival paths for adversaries that can lead to cyber-physical attacks amongst other reliability issues in the communication layer. This long-standing problem necessitates new intrinsic ways of exchanging information between converters through power lines to optimize the system's control performance. Going beyond the existing power and data co-transfer technologies that are limited by efficiency and scalability concerns, this paper proposes neuromorphic learning to implant communicative features using spiking neural networks (SNNs) at each node, which is trained collaboratively in an online manner simply using the power exchanges between the nodes. As opposed to the conventional neuromorphic sensors that operate with spiking signals, we employ an event-driven selective process to collect sparse data for training of SNNs. Finally, its multi-fold effectiveness and reliable performance is validated under simulation conditions with different microgrid topologies and components to establish a new direction in the sense-actuate-compute cycle for power electronic dominated grids and microgrids.

What problem does this paper attempt to address?

This paper discusses a new communication method called Neuromorphic Event-Driven Semantic Communication in microgrids. The coordination control of traditional microgrids relies on communication networks, which introduces data leakage paths and possible network security attacks. The paper proposes a novel communication strategy based on Spiking Neural Networks (SNNs) to directly exchange information between nodes using power lines in order to optimize system control performance. Existing power and data co-transmission technologies, such as Talkative Power Communication (TPC), suffer from efficiency and scalability issues and are constrained by the voltage levels and topology structures of power systems. The paper suggests using SNNs to collect sparse data through event-driven selective processes and train the SNNs of each node in an online collaborative manner. This approach reduces dependency on traditional communication infrastructure and improves system robustness and efficiency. Through simulations, the effectiveness and reliability of this method are demonstrated under different microgrid configurations, providing a new direction for the perception-action-computation cycle in power electronics-dominated grids and microgrids. The main contributions of this research are: 1) utilizing neuromorphic computing for coordinated control in microgrids, 2) using event-driven SNNs for online learning using only power exchange data between nodes, and 3) proposing a communication principle that is low-energy, low-latency, and not limited by physical isolation stages. The paper also discusses the comparison with traditional communication methods such as TPC and traditional cyber-layer communication, as well as the advantages of SNNs in terms of energy efficiency, latency, and hardware performance. Finally, the paper describes the basic operation principles of SNNs and their applications in microgrids, including data collection, offline design, and the process of event transformation into spike signals.

Neuromorphic Event-Driven Semantic Communication in Microgrids

Inferring Ingrained Remote Information in AC Power Flows Using Neuromorphic Modality Regime

Sensorless Control of DC Microgrid Based on Artificial Intelligence

Spike Talk in Power Electronic Grids -- Leveraging Post Moore's Computing Laws

Spike Talk: Genesis and Neural Coding Scheme Translations

Power Talk: A novel power line communication in DC MicroGrid

Spiking Neural Networks -- Part III: Neuromorphic Communications

Unsupervised Learning for Non-intrusive Load Monitoring in Smart Grid Based on Spiking Deep Neural Network

Neuromorphic Near-Sensor Computing: From Event-Based Sensing to Edge Learning

Cellular computational generalized neuron network for frequency situational intelligence in a multi-machine power system

Distributed and Asynchronous Operational Optimization of Networked Microgrids

Coordinating Multi-Energy Microgrids for Integrated Energy System Resilience: A Multi-Task Learning Approach

Compiling Spiking Neural Networks to Neuromorphic Hardware

NeurJSCC Enabled Semantic Communications: Paradigms, Applications, and Potentials

Toward a Cyber Resilient and Secure Microgrid Using Software-Defined Networking

Integration of neuromorphic AI in event-driven distributed digitized systems: Concepts and research directions

Resilient Communication Scheme for Distributed Decision of InterconnectingNetworks of Microgrids

A Resilience-Oriented Centralised-to-Decentralised Framework for Networked Microgrids Management

Agent-driven Generative Semantic Communication with Cross-Modality and Prediction

SNN-SC: A Spiking Semantic Communication Framework for Collaborative Intelligence

Reducing Detrimental Communication Failure Impacts in Microgrids by Using Deep Learning Techniques