|
Abstract
|
The increasing penetration of renewables in distribution grids poses a complex planning problem for network operators. Renewable generation (e.g., wind or solar) is inherently intermittent, and real-time load demand is ever-changing, so the system operates under significant uncertainty. ESSs can mitigate these issues by buffering fluctuations, smoothing the load/generation balance, and supporting voltage and frequency control—thereby improving overall power quality and reliability. However, deciding how to optimally locate, size, and operate an ESS in the network is non-trivial because it must account for a wide range of operating scenarios and balance multiple (often conflicting) performance goals. Technically, the ESS planning problem is multi-dimensional: the planner must minimize economic costs (energy losses and outage costs) while maintaining acceptable voltage profiles and stability margins across all buses. Solving this is challenging due to the non-linear AC power flow constraints, the mix of discrete decisions (whether and where to install storage) and continuous variables (dispatch levels), and a 24-hour operational horizon with stochastic inputs. Traditional planning strategies that ignore uncertainty or optimize only a single objective (e.g., cost minimization) are inadequate—a solution that appears least-cost may lead to undervoltage issues or reduced stability if voltage criteria are neglected. In particular, failing to consider voltage stability in planning can be dangerous: if an ESS placement drives the system closer to its stability limit, a small change in load could trigger large voltage swings or even voltage collapse. Moreover, modeling many scenarios and multiple objectives introduces high dimensionality that makes conventional optimization techniques computationally infeasible or prone to getting trapped in suboptimal solutions. This necessitates the use of advanced heuristic algorithms that can efficiently search the solution space and avoi
|