INFORMATIC STUDY AND FEEDFORWARD CONTROL METHOD OF SYSTEM STATE DYNAMICS

点击翻译按钮获取中文摘要

My old theory' state transformation' has been now developed into the new theory 'uncertain state system superimposed upon a general system'. Then, its characteristic value, namely system information, is defined by the difference between the information ehtropy of the combined distribution and that of the joint distribution. These viewpoints proposed some new fields in this research. For example, when an uncertain state system is given by the vectors x, y, z---etc., the system information is as follows.S (x ; y ; z ; ---) =H (x・y・z---) -H (xyz---)Here is an important remark ; the system information could be decomposed into lower dimensional components according to our problem. If the system is three-dimensional, its information is expressed by S (x ; y ; z). In this case, the system is decomposed into three kinds of two-dimensional systems.There are three kinds of system information ; S (x ; y), S (y ; z) and S (y ; z). Of course, the uncertain state system must have the complementary branch structure expressed in the former research reports. Those ideas are very important in this system theory.As for an application example, there is the problem of information when the table of three items is changed into the three tables of two items (i. e. the one from three-dimensional to two-dimensional). If the following inequality is found,S (x ; y ; z)*S (x ; y) + S (y ; z) +S (y ; z).the loss of information may occur by the transformation of expression form. Moreover, the above mathematical approach is expected to be helpful in the field of data mining.