基于ANN的变电站虚拟急救操作有效性预测方法

An effectiveness prediction approach for virtual first-aid operation based on ANN

为了提升虚拟急救培训的有效性,结合变电站工作的特殊性,提出一种基于人工神经网络(Artificial Neural Network, ANN)的变电站虚拟急救操作有效性预测方法。首先,将变电站相关身体状态指标加入小样本急救数据集,采用蒙特卡洛方法对数据集进行扩充;其次,构造并训练ANN模型,预测急救操作与患者身体状态变化之间的关联;最后,开发了一套多维度急救数据可视化系统,实时展示急救操作效果,指导学员调整急救操作。实验结果表明,提出的方法可以有效预测施救效果,为特殊场景下的虚拟急救培训提供直观指导。

To improve the effectiveness of virtual first aid training, a smart effectiveness prediction approach for virtual first aid operation based on the Artificial Neural Network(ANN) is proposed, considering the particularity of working in transformer substations. Firstly, the substation-related physical state index is added to the small sample first aid data set, and the data set is expanded by Monte Carlo method. Secondly, an ANN model is constructed and trained to predict the correlation between the first aid operations and changes in the patient’s physical state. Finally, a multi-dimensional first aid data visualization system is developed to show the effect of first aid operation in real time and guide students to adjust first aid operation. Experimental results show that the proposed method can effectively predict the rescue effect, and provide intuitive guidance for virtual first aid training in special scenes.