Architecting smart electrical appliance network for self-approaching optimization operation

Developing specialized agents for electrical appliances and integrating them into a Smart Electrical Appliance Network (SEAN) has emerged as a primary method for smart electricity usage applications. However, electricity consumer behavior and environmental conditions exhibit temporal variability, posing significant challenges to accurately modeling consumer satisfaction with electricity usage. These factors contribute to the complexity of optimizing methods to deliver satisfactory electricity services and management. In this paper, we introduce a method that enables SEAN to achieve self-approaching optimization operation, aiming to deliver electricity services and management that meet consumer satisfaction. Initially, leveraging an autonomous decentralized system, we designed the architecture of SEAN, including the types and functionalities of specialized agents, data-driven approaches, data fields, and content codes. Subsequently, we implemented the autonomous control processor for each specialized agent using behavior trees, which showcased high reactivity and modularity in handling complex internal relationships involving application programming, data acquisition and broadcasting, as well as event monitoring. Finally, in a practical implementation, a SEAN comprising 16 specialized agents was used to validate the effectiveness of the proposed electricity services and management methodologies. The results demonstrate that our approach ensures consumer satisfaction from the bottom up, while also providing high-quality electricity services and management. Additionally, our SEAN solution exhibits advantages in reducing CPU usage, minimizing data traffic, and shortening development cycles, further underscoring its utility and efficiency in smart electricity management contexts. ● Developed agents for electrical appliances, establishing a Smart Electrical Appliance Network. ● Utilized an Autonomous Decentralized System to design the architecture of the Smart Electrical Appliance Network. ● Behavior Tree is employed to implement autonomous control processors for agents. ● Demonstrated the functionality and viability through instantiation in a university laboratory.