摘要
协同式专家系统的任务分解,指将总任务合理地分解成几个分任务,分配给子专家系统,分别由子专家系统进行任务求解。各子专家系统相互协作、共同解决一个相对复杂的问题。该文主要采用两种算法,黑板模型法和与或树法,在遵循任务分解原则的基础上,对电站调试的任务进行分解,分解后的任务粒度要适宜,规模适中,均衡,易于单个子专家系统解决,同时既要考虑局部优化,又要使全局最优,不能因为分解后的任务负载过重,而使子专家系统无法解决。
Coorperative expert system task decomposition, refers to the general task reasonably broken down into several sub - tasks, assigned to several sub - expert system, experts from several hours to complete the system. So that each sub -sub -expert system for mutual cooperation, together to resolve the issue of a broader field. In this paper, two algorithms, blackboard model law and with or tree method, in complying with the principle of task decomposition, of power stations break down the task of debugging. Decomposed to fit the task granularity, moderate scale, balance, a single sub - expert system is easy to solve. At the same time, it is necessary to consider local optimization, but also so that the global optimum, can not decompose the task after the overload, sub - expert system which could not be resolved.
出处
《工业仪表与自动化装置》
2009年第5期68-70,73,共4页
Industrial Instrumentation & Automation
关键词
协同式
任务分解
粒度
消耗
cooperative
task decomposition
size
consumption
