基于预警文本信息的调度命令智能生成模型

Intelligent Generation Model of Dispatching Command Based on Early- warning Text Information

摘要:

提出调度命令智能生成模型,该模型由神经网络和调度命令修正模块两部分组成。基于长短时记忆（LSTM)网络构建序列?序列（seq2seq）模型，将预警文本信息作为模型的输入进行训练，提出五种调度命令修正策略并分别对五种易错信息进行修正，最终得到调度命令。结果表明：该模型具有利用预警文本信息生成调度命令的能力，引入的调度命令修正模块能够有效提升调度命令生成质量。

Abstract:

An intelligent generation model of dispatching commands is proposed in this paper. The model consists of a neural network and a dispatching command modification module. A sequence-sequence (seq2seq) model based on a long-term and short-term memory（LSTM) network is built, the training is performed using early-warning text information as input into the model. Five scheduling command modification strategies are put forword and the error-prone information is modified to obtain the final scheduling command. It is shown that the model has the ability to generate dispatching commands by using early-warning text information. The introduction of a command correction module can effectively improve the quality of dispatching command generation.