Abstract
<jats:p>The widespread deployment of residential battery energy storage systems (BESS) introduces complex control challenges in tropical low-voltage distribution networks, where high ambient temperatures accelerate transformer degradation. This study presents a four-layer cyber-physical transactive energy co-simulation framework for coordinating a decentralized fleet of 55 residential BESS assets. The architecture integrates a double-auction market driven by thermal- and voltage-sensitive distribution locational marginal pricing (DLMP) benchmarked against an omniscient, centralized non-linear programming (NLP) optimization solver. Modeled on an unbalanced Ghanaian network, the framework mitigates voltage unbalance factor (VUF) violations and transformer insulation wear. Quantitative results show that while uncoordinated baselines and rigid centralized overrides fail to prevent severe thermal core stagnation (0.0198 h daily aging loss), the proposed predictive transactive framework reduces diurnal insulation degradation to 0.0184 h (7.1% asset life extension) while preserving data privacy. In comparison, the centralized NLP benchmark established a global theoretical upper bound, restricting the aging loss to 0.0155 h (21.7% increase). This study validates that market-driven, incentive-based hierarchical control can help reconcile decentralized prosumer autonomy, network phase balancing, and utility infrastructure longevity under stringent operating constraints.</jats:p>