考虑计算资源动态聚合的算力中心微电网协调调度优化

Abstract The computility center (CC) is a flexible load that regulates its power demands through workload dispatching. CC microgrid can employ the flexibility of CC load to coordinate with the volatility of photovoltaic (PV) generation. However, the computing resources of CC are heavily fragmented. Such situation limits workload distribution and then results in CC load incapable of coordinating with microgrid. To address the above issue, this paper proposes a coordinated scheduling method for CC microgrid using computing resources dynamic pooling (CRDP). Specifically, a workload-core mapping model is proposed to transform workload into power load by formulating the matrix of processor core states. Subsequently, the CRDP method is proposed to integrate and allocate remaining available cores according to the core real-time states. Then a self-regulating CC microgrid coordinated framework is proposed to regulate the CC load by adjusting the scale of computing resource pool to the volatility of PV generation. The effectiveness of the proposed method in improving PV consumption is validated across different microgrid simulation scenarios.