轨交杂散电流对燃气管道影响预测的LSTM模型

LSTM Model for Prediction of Influence of Rail Transit Stray Current on Gas Pipeline

根据长短期记忆网络(LSTM)具有较好的数据连贯性及相关性、处理时序性问题时准确度较高的特点,建立轨交杂散电流对燃气管道影响预测的LSTM模型,该模型建模简单,易于操作。对遭受杂散电流干扰的埋地钢质燃气管道通电电位在同一位置进行持续监测,应用LSTM模型对杂散电流作用于埋地钢质燃气管道的影响进行了短期预测,并与实测结果进行了对比分析,准确度较高。利用LSTM模型对受到轨交杂散电流影响的燃气管道状况作出预测,可以为研制开发更智能的具有反馈功能的排流装置提供支撑和参考。

According to the characteristics of long short-term memory(LSTM)network that has better data coherence and relevance,and high accuracy when dealing with temporal problems,a LSTM model for predicting the influence of rail transit stray current on gas pipeline is established.The model is simple and easy to operate.The energization potential of buried steel gas pipeline interfered by stray current is continuously monitored at the same location.The short-term prediction of the effect of stray current on buried steel gas pipeline is carried out by using LSTM model,the results are compared with the measured results,and the accuracy is high.The LSTM model is used to predict the conditions of gas pipeline affected by rail transit stray current,which can provide support and reference for the development of more intelligent drainage device with feedback functions.