一种新型智能监控系统

设计了一种具有学习功能的多点实时控制的智能监控系统,将模糊小脑模型、多层前馈网参数分离算法用于学习模块,增大了系统经验值,并提高了学习速度;用信息融合模型解决局部与整体的关系,用多信道对同一点监控及多点并行监控多目标的方法,解决了误报、漏报、监控范围小、报警信息量单一等问题,并以多信道对同一点监控为例,进行了仿真验证.

A Study on a New Type of Intelligence Monitor System

A new type of time-control intelligence monitor system, which has a learning function and can control several time points simultaneously, is introduced. It applies misty cerebellum model and the parameter separation algorithm of multiple-layered forward feedback net to learning template so as to improve the system experience value and learning speed. It uses information integration model to deal with the relationship between the part and the whole and monitors one or several time points by using the multi-information passage. This new system solves the problems of wrong alarms, alarm slips, the narrowness of monitoring and information monotony of alarms. An experiment on multi-information passage monitoring the same time point is conducted to simulate the system.