The Self-Study and Auto-Control Model of Machine´s Lifetime Learning Based on Database

The paper presents a novel design methodology for an intelligent controller based on a database capable of self-study and auto-control. The motivation stems from making a smart database analogous to human brain, namely, a dummy brain database (DBD). The data stored in the DBD may be collected from a plant/system at present, recent past and/or over a period of time in the past, i.e. the historical data. The proposed method intends to enable DBD to perform information processing, knowledge elicitation and decision making in a manner similar to human brain. In particular the features of human memory including its sensitivity, frequency of change, information update and reconstruction are researched and the mechanism of the brain´s associative memories studied. The DBD thus established has led to a self-study and auto-control model which is capable for learning over the machine´s lifetime. Possible applications in control of a group of elevators and in city traffic control are discussed