Real-Time Power System State Estimation and Forecasting via Deep Unrolled Neural Networks
Real-Time Power System State Estimation and Forecasting via Deep Unrolled Neural Networks
复制标题
DOI：
10.1109/tsp.2019.2926023
复制
发表时间：
2018-11
影响因子：
5.4
通讯作者：
Liang Zhang;G. Wang;G. Giannakis
中科院分区：
文献类型：
--
作者：
Liang Zhang;G. Wang;G. Giannakis
当代电网正面临因大规模部署可再生发电机、电动汽车和需求响应计划而导致的快速且大幅电压波动的挑战。在此背景下，实时监控电网运行状况变得越来越重要。随着大规模和非凸性的出现，现有的电力系统状态估计（PSSE）方案的计算成本变得昂贵，或者常常产生次优的性能。为了绕过这些障碍，本文提倡使用受物理启发的深度神经网络（DNN）进行实时电力系统监控。通过展开最初使用精确 ac 模型开发的迭代求解器，为实时 PSSE 开发了一种新颖的特定于模型的 DNN，仅需要离线训练和最少的调整工作。为了进一步实现系统感知，甚至在时间范围之前，以及赋予基于 DNN 的估计器弹性，深度循环神经网络 (RNN) 也被用于电力系统状态预测。深度 RNN 利用历史电压时间序列中存在的长期非线性依赖性来实现预测，并且易于实现。数值测试表明，与现有替代方案相比，所提出的基于 DNN 的估计和预测方法的性能得到了提高。在 IEEE 118 总线基准系统的实际负载数据实验中，基于特定模型的 DNN PSSE 方案的性能几乎优于其竞争方案（包括广泛采用的高斯牛顿 PSSE 求解器）一个数量级。
Contemporary power grids are being challenged by rapid and sizeable voltage fluctuations that are caused by large-scale deployment of renewable generators, electric vehicles, and demand response programs. In this context, monitoring the grid's operating conditions in real time becomes increasingly critical. With the emergent large scale and nonconvexity, existing power system state estimation (PSSE) schemes become computationally expensive or often yield suboptimal performance. To bypass these hurdles, this paper advocates physics-inspired deep neural networks (DNNs) for real-time power system monitoring. By unrolling an iterative solver that was originally developed using the exact ac model, a novel model-specific DNN is developed for real-time PSSE requiring only offline training and minimal tuning effort. To further enable system awareness, even ahead of the time horizon, as well as to endow the DNN-based estimator with resilience, deep recurrent neural networks (RNNs) are also pursued for power system state forecasting. Deep RNNs leverage the long-term nonlinear dependencies present in the historical voltage time series to enable forecasting, and they are easy to implement. Numerical tests showcase improved performance of the proposed DNN-based estimation and forecasting approaches compared with existing alternatives. In real load data experiments on the IEEE 118-bus benchmark system, the novel model-specific DNN-based PSSE scheme outperforms nearly by an order-of-magnitude its competing alternatives, including the widely adopted Gauss–Newton PSSE solver.