高通量筛选系统的双子代数建模

Dioid-based modeling of high throughput screening systems

对一类新兴的离散事件系统—–高通量筛选系统而言,max-plus算法是强有力的事件域建模与分析工具,因此,其常被用来进行该系统的建模及控制.然而,有时,尤其是在执行控制策略的时候,需要时间域的描述.为了同时从时间域和事件域高效描述系统,本文利用基于双变量的形式幂级数,把max-plus算法模型扩展为双子代数模型.对可能产生的非因果模型,为便于进行系统分析、仿真和后续的控制,提出了多个相应的基于双变量(时间变量和事件变量)的转换矩阵.通过矩阵转换,得到满足要求的因果模型.最后用实例表明了该方法及其在提高该系统控制效率方面的有效性.

For a new class of discrete event systems — high throughput screening systems(HTS), max-plus algebra is a powerful event domain modeling and analysis tool and therefore, is often used for its modeling and control. However, time domain description is sometimes also preferred, especially when it comes to the implementation of HTS control strategies.To efficiently describe the systems in both time domain and event domain, max-plus algebra models are extended to dioid models with the help of two-variable formal power series. For a possible resulted non-causal model, in order to facilitate the system analysis, simulation and the following control, the corresponding two-variable(time variable and event variable)transformation matrices are proposed. The desired causal model could be derived from matrix-transforming. An example is given to demonstrate the method and its effectiveness in enhancing the control efficiency of HTS.