箱式变电站由于其灵活性和适应性在电力系统改扩建过程中起着重要作用，由于常置于户外，环境温湿度对其工作状况具有重要的影响。针对现有箱式变电站针对温度和湿度直接控制存在的调整滞后等问题，文中提出了基于温湿度变化率预测的调节控制方法，首先论述了基于温湿度预测的调控系统结构，然后提出了基于长短期记忆 (long-short term memory, LSTM)人工神经网络模型的箱式变电站温湿度变化率预测方法和基于变化率预测结果的调节模式，最后，通过仿真实验分析验证了所述方法在箱式变电站环境温湿度调控中的有效性。
英文摘要:
Box-type substations play an important role in renovation and expansion of power system due to their flexibility and adaptability. As they are usually placed outdoors, environmental temperature and humidity have a significant impact on their working conditions. Aiming at the problems of adjustment lag in the direct control of the temperature and humidity control issues in the existing box-type substations, a regulation and control method based on temperature and humidity changing rate prediction is proposed in this paper. Firstly, the regulation and control system structure based on temperature and humidity prediction is discussed, and then, a temperature and humidity changing rate prediction method based on long short-term memory(LSTM) model and a regulation mode based on changing rate prediction results are proposed for box-type substations. Finally, the effectiveness of the proposed method in temperature and humidity control of box-type substations is verified through the analysis of simulation experiments.