English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Power system optimal dispatch with transient security constraints is commonlyrepresented as Transient Security-Constrained Optimal Power Flow (TSC-OPF).Deep Reinforcement Learning (DRL)-based TSC-OPF trains efficientdecision-making agents that are adaptable to various scenarios and providesolution results quickly. However, due to the high dimensionality of the statespace and action spaces, as well as the non-smoothness of dynamic constraints,existing DRL-based TSC-OPF solution methods face a significant challenge of thesparse reward problem, which greatly limits the practical applications of thesemethods in large-scale power grids. To address this issue, a fast-converged DRLmethod for TSC-OPF is proposed in this paper. On the one hand, the MarkovDecision Process (MDP) modeling of TSC-OPF is improved by reducing theobservation space and smoothing the reward design, thus facilitating agenttraining. On the other hand, an improved Deep Deterministic Policy Gradientalgorithm with Curriculum learning, Parallel exploration, and Ensembledecision-making (DDPG-CPEn) is introduced to drastically enhance the efficiencyof agent training and the accuracy of decision-making. The effectiveness of theproposed method is demonstrated through experiments on the IEEE 39-bus systemand a practical 710-bus regional power grid. In both systems, the trainingprocess shows stable and rapid convergence. Furthermore, the efficiency andaccuracy of the well-trained agents are confirmed by testing them in newscenarios and comparing their dispatch results with those obtained using theparticle swarm optimization algorithm. The source code of the proposed methodis made public on GitHub.

On Fast-Converged Deep Reinforcement Learning for Optimal Dispatch of Large-Scale Power Systems under Transient Security Constraints