FAlloc: A Fair Power Limit Allocation-Based Approach to Implement Brownout

• Published in: Journal of control, automation & electrical systems Vol. 35; no. 2; pp. 361 - 375
• Main Author: Agarwal, Anshul
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.04.2024; Springer Nature B.V
• Subjects: Algorithms; Blackouts; Brownouts; Circuit breakers; Control; Control and Systems Theory; Datasets; Demand side management; Electric utilities; Electrical Engineering; Electricity; Energy; Engineering; Households; Linear programming; Mechatronics; Optimization; Power failures; Power supply; Robotics; Robotics and Automation; Scheduling; Thresholds; Demand side management; Smart metre; Blackout; Peak power demand; Brownout; Smart grid; Distribution algorithms
• ISSN: 2195-3880; 2195-3899
• DOI: 10.1007/s40313-024-01077-x

Due to the significant gap between demand and supply of electricity, power utility companies employ rolling blackouts, in which the power supply to different regions is cut-off periodically for a specified duration. However, this causes increased user inconvenience as consumers are left without power; also, a large undesirable blackout may occur if the circuit breakers are not opened in time. To solve this problem, this paper has developed a more pragmatic approach in which a power threshold limit is imposed on households during times of deficit power supply. This paper presents five algorithms designed to allocate load thresholds to households in a fair manner. Some of these algorithms perform the worst (high violations) in some time slots and best (minimal violations) in other slots, but none guarantees minimal mean violation across all the time slots. In this regard, this paper has also developed a novel optimal algorithm that ensures a minimal violation percentage of the allotted thresholds across all time slots in all the households of a neighbourhood. It also employs multiple heuristics for threshold allocation, thereby preventing household starvation. The algorithms are evaluated on a real-world dataset. Its mean violation across all the houses goes as low as nearly 45% and is minimum across all the time slots. It performs all the algorithms used for power limit distribution. This result demonstrates the effectiveness of the developed technique in the implementation of the brownout scheme in practise.