基于离散Hopfield神经网络的化学实验室安全评估

Chemical Laboratory Safety Evaluation Based on DiscreteHopfield Neural Network

针对高校化学实验室安全风险难以量化评估的问题,采用一种基于离散Hopfield神经网络(discrete Hopfield neural network,DHNN)的化学实验室安全评估方法.首先,利用层次分析法建立化学实验室安全状况多指标评估体系;然后,使用模糊综合评价法对评估指标进行量化,对评估指标编码;最后,使用学习率对DHNN进行优化,将该方法与传统评估方法进行对比,结果表明该方法能够实现对样本的准确评估.将该方法应用于高校危险化学品实验室安全评估过程中,仿真实验结果表明该方法构建的指标体系合理可行且评估精度较高.

To solve the issue of evaluating the safety risk of chemical laboratories in universities,a safety evaluation method based on discrete Hopfield neural network(DHNN)was adopted.First,a multi-index safety status evaluation system was established by using analytic hierarchy process.Then,the fuzzy comprehensive appraisal was used to quantify the evaluation indexes and encode evaluation indicators.Finally,an optimization algorithm based on learning rate was adopted.This method was compared with traditional evaluation methods and the results show that this method is capable of achieving an accurate evaluation of the sample.After applying this method to real scenarios,simulation results demonstrate that the index system is reasonable and feasible,and the evaluation accuracy is high,which can provide a reference for practical safety risk evaluation.