基于深度神经网络的电力调度稳定控制方法

Power Dispatching Stability Control Method Based on Deep Neural Network

针对现有控制方法在对电力调度控制时,存在机电设备实部稳定概率和阻尼比稳定概率过低,无法保证其运行安全和稳定的问题,引入深度神经网络,开展对其稳定控制方法的研究。结合自回归滑动平均函数,在控制体系中引入自适应广域阻尼控制器;利用深度神经网络,完成对稳定控制模型的构建和训练;采用实时决策方式,实现电力调度的实时稳定控制。通过对比实验证明,新的控制方法应用下,电力系统中各个机电设备的实部稳定概率和阻尼比稳定概率得到显著提升,能够在极大程度上为电力调度的稳定控制提供重要保障。

In view of the problem that the stability probability and damping ratio of mechanical and electrical equipment are too low in the existing control methods for power dispatching control,which can not ensure the safety and stability of its operation,deep neural network is introduced to carry out the research on its stability control method.Combined with the autoregressive moving average function,an adaptive wide-area damping controller is introduced into the control system.The stability control model is constructed and trained by using deep neural network.Real-time decision-making is adopted to realize real-time and stable control of power dispatching.Through comparative experiments,it is proved that the real stability probability and damping ratio stability probability of each mechanical and electrical equipment in the power system are significantly improved under the application of the new control method,which can provide an important guarantee for the stability control of power dispatching to a great extent.