Researchers from China’s State Grid Henan Electric Power Company have proposed a Stackelberg-game-based optimal allocation method for distributed energy storage systems (DESSs). A Stackelberg game is a decision-making model in which one participant – the leader – acts first, while other participants – the followers – respond to that decision.
“The optimal allocation of DESSs is a high-dimensional, nonlinear, and complex optimization problem. Existing methods struggle to obtain high-quality solutions within a limited time, and most fail to consider the game behaviors among multiple stakeholders fully,” explained the research group. “The multi-leader multi-follower Stackelberg game model is applied for the first time to the optimal allocation of DESSs, overcoming the limitation of traditional centralized optimization methods.”
The novel technique first analyzes grid demand and renewable-energy fluctuations, separating fast and slow changes so the most suitable storage technologies can handle them. It then models the electricity market as a Stackelberg game, with energy suppliers setting prices, and storage operators and consumers responding by adjusting their charging, discharging, and electricity use.
Then, a Pareto-archival multi-objective particle swarm optimization (PAMOPSO) algorithm repeatedly tests different storage locations, capacities, prices, and operating schedules. While doing so, it also rejects options that violate cost, voltage, power balance, safety, or battery limits. The process continues until it identifies stable solutions that balance the interests of all participants while improving grid performance and reducing the amount of storage required.
The new framework was demonstrated through a computer-simulated case study of a 12.66 kV, 33-node distribution network, using 90 days of real load and solar-generation data recorded at 15-minute intervals under sunny, cloudy, and rainy conditions. The network included two 200 kW solar systems, while the algorithm was allowed to select up to two storage locations and determine their capacities and operating schedules.
Experimental results show that, in a standard distribution network test case, the load fluctuation rate is reduced from 25.98% to 14.25%, representing a reduction of 11.73%. “The required energy storage capacity is below 500 kWh, approximately 85% lower than that of traditional methods,” the team said.
The average total cost of the optimized distributed energy-storage configuration over the five-year planning period was CNY 1.8187 million ($269.436), according to the results, which was lower than the reference method tested. The constraint satisfaction rate reached 99.8%.
The results were presented in “Optimal allocation of distributed energy storage systems based on the Stackelberg game,” published in Results in Engineering.
The post Stackelberg game framework optimizes distributed energy storage allocation appeared first on Energy Storage.
