基于熵的数字化人机交互复杂度研究  

Research on digital human-machine interaction complexity based on entropy

作　　者：张力[1,2] 刘雪阳[1] 洪俊[2] 胡鸿[2] 邹萍萍[1] 

机构地区：[1]南华大学人因研究所,湖南衡阳421001 [2]湖南工学院人因工程与安全管理研究所,湖南衡阳421002

出　　处：《中国安全科学学报》2015年第10期65-70,共6页China Safety Science Journal

基　　金：国家自然科学基金资助(71071051;71371070);湖南省战略性新兴产业产学研结合创新平台创新能力建设项目(2012GK4101)

摘　　要：为预防数字化人机交互系统的人因失误,建立一种分析人机交互复杂度(HMIC)的新方法,并进行实践验证。首先,用信息熵方法研究数字化HMIC。基于数字化控制系统中人机交互与传统系统中的区别,将任务逻辑复杂度(TC)、操作步骤复杂度(OC)、信息复杂度(IC)与知识水平复杂度(KC)确定为度量数字化人机交互复杂性的主要指标,且集成为一个综合性指标-HMIC。然后,基于熵值法建立指标权重调整模型,使静态赋权和动态赋权相结合,增强评价的合理性和科学性。通过欧几里得范数得到HMIC计算模型。用该模型,分析某实际数字化核电厂蒸气发生器传热管破裂(SGTR)应急操作过程的HMIC。结果表明,基于熵的数字化HMIC模型和方法是合理、可行的,适合在工程中应用。It was held by the authors that in order to prevent human-caused human-machine interaction system failure,a new method should be worked out and verified for analyzing digital human-machine interaction system complexity（ HMIC）. Information entropy was used for analyzing the digital HMIC. Based on differences in human-machine interaction between a digital system and a conventional system,the task logical complexity（ TC）,operation-step complexity（ OC）,information complexity（ IC） and knowledge-level complexity（ KC） were identified as four main indexes for evaluating a comprehensive measuring index：HMIC. A index weight adjustment model was built based on information entropy,which enhances rationality and scientificity of the evaluation by combining static weight with dynamic weight. An HMIC calculation model was built on the basis of Euclidean norm. Using the IC calculation model,the IC of the steam generator to be rupture（ SGTR） procedure of a certain nuclear power plant in China was analyzed quantitatively. The experimental results show that this model and method are reasonable and feasible,and it can be applied in engineering.

关 键 词：熵 数字化 人机交互复杂度(HMIC) 权重 蒸气发生器传热管破裂(SGTR) 

分 类 号：X912.9[环境科学与工程—安全科学]