Research In Intelligent Control Technologies Rough Set-Based In Power Drives

This paper introduces the meaning of the research on power drive control system, analyses the development of intelligent control, and puts forward intelligent control technology based on rough set apply to power drive system.Rough set is a new soft computing method. This paper analyzes idea pith and characteristic of rough set theory deeply, introduces its combination with other soft computing methods such as fuzzy sets, neural networks, evolution algorithms, and then summarizes its current application in intelligent control systems.Attribute reduction is a core of rough set theory, and many attributes reduction algorithms were put forward. The author makes an improvement based on former achievement in this paper. It gives a new definition of information quantity, and put forwards new attribute reduction arithmetic that using relative core as jumping-off point of attribute reduction, and attribute importance as heuristic information. It can reduce searching space and time during knowledge reduction.Because rough set theory is the analysis of discrete finite-dimensional data tables, we need to discretize and normalize the data when mining data by rough set theory. But for an original database with large continuous attributes values, how to discretize data reasonably is a noticeable problem. Aiming at this condition, base on the characteristics of mind evolutionary algorithm which can find the best point globally, we put forward global discrete method that considering the whole condition attributes values, and search reasonable split points. Thereby it can avoid bringing unreasonable discrete split points when local discretizing single condition attributes values, then avoid bringing differ rule. By using the method in this paper; the input and output data of the typical power drives systerm--boiler are analyzed, and the fuzzy model of boiler is established. The experiment results show that the method is simple and feasible.Applying this method to the double closed-loop DC regulating velocity system for thyristor, and intelligent control strategy based on rough set and mind evolutionary algorithm is presented in this paper, which integrate the superiority of rough sets in data analysis and the capability of mind evolutionary algorithm which can find the best point globally, associate with fuzzy control. The experiment results show that this control strategy can improve some control quality, the fuzzy control rules can reflect control performance of the control system.