AI-driven metaheuristic optimization for enhanced grid resilience under extreme weather and load shedding constraints

Engineering notes

Engineering notes will be added by the Power for Solar editorial team.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为光伏效率、钙钛矿太阳能电池、储能技术、太阳能热利用、BIPV、并网技术等高价值论文补充中文说明。

Original abstract

Abstract The incorporation of Distributed Energy Resources (DER), especially PV and BESS, is essential in making radial power systems stable. Unfortunately, most recent optimization algorithms make an assumption of an unconstrained utility supply, targeting minimization of losses during daylight or only profit-making energy trading. Such traditional methods will not work in underdeveloped areas that face significant capacity limitations in their main substations, forcing them to undertake mandatory load shedding. In such situations, having an unconstrained BESS leads to parasitic voltage drop at weak tail nodes, whereas reactive disconnect switches lead to excessive Energy Not Served (ENS) production. In order to fill this gap, an artificial intelligence-based approach of survivability control that changes the focus from economic dispatching to active grid survivability has been proposed in this study. By using Particle Swarm Optimization (PSO), the proposed approach incorporates modelling of artificial solar intermittency, constrained and asymmetric dispatching of BESS systems, and a demand-side management procedure. A penalization technique is used to ensure strict adherence to the statutory voltage constraints. Using the IEEE 33 bus network system with the maximum capacity for active power at 4.0 MW, the AI-based approach performed significantly better compared to the conventional distributed generation strategies. The new proposed approach was able to ensure that there were no violations of the absolute minimum network voltage (0.95 p.u.), reduced power losses to 1.696 MWh (a 31.6% improvement over static DG methodologies), and prevented all instances of unexpected Energy Not Served (ENS).

Links and sources

• PDF from original source

Comments