关键词: 非统计分析, 分布式复杂机电系统, 分布式控制系统数据, 色彩图谱, 数据驱动

Abstract: Distributed control system (DCS) is a subsystem of the on-line monitor system in modern industry, which responses for collect runtime data in order to realize on-line monitor of the whole process. It is invalid for all traditional methods to get the health state of the system because of the large-scale and higher dimensional of DCS data. Hence, it is still in the primitive state, which is single point monitor state. In order to solve this problem, we define a homeomorphism product space of DCS data that named color phase space which is a Cartesian product of two-dimensional Euclidean space and three-dimensional color phase space. So mapping the DCS data to the color phase space can get a system color picture which represents the internal relationship of variables of the complex system. Hence, the system color picture reflects the relationship of nonlinear and high coupling of multiple variables of the system from the macro level of the whole system according with the method of half-qualitative and half-quantitative. It solves the problem of establish analytical relationship of multiple variables of the complex system and makes the problem become visual and simple.

Key words: Color picture, Data drive, Distributed and complex electromechanical system, Distributed control system (DCS) data, No-statistical analysis