Research On Intelligent Control System Based On Rough Set Theory

Rough set theory is a new mathematical and artificial intelligent technique in knowledge discovery and is becoming the basic theory in machine learning. Rough set can derive control rules from process data automatically and the control algorithm can depend on process data completely, so rough set has great advantages and potential in rule-based control system. This paper does some research on the application of rough set in control field surrounding how to extracte rules and build rough controller.Discretization of continuous attributes is the key step in processing continuous problem with rough set theory. This paper presents an algorithm for discretization based on binary particle swarm optimization (BPSO) and a new bound cuts strategy is proposed to decrease the num of candidate cuts so as to reduce the complication of algorithm. Then the fitness is on the promise of keeping the consistency of information system and the less number of cut points. And at the same time this paper makes an experiment on several datasets from UCI Machine Learing Repository by this method which has proved that using BPSO to solve discretization of rough set is efficientlly.In order to validate the efficient of rules extracted from input-output datas based on rough set and explain the design method of rough controller, this paper constructs the rough control system based on PID controller. This method uses extra testing signals to excite the process for acquiring sufficient data that can reflect the control laws of the operator or the existing controller and use the method of Continuous Attribute Whole Distance (CAWD) to decrese universe space dimensions which can reduce the latter workload. A satisfying control performance of the rough controller can be achieved by simulation. Then settle the adaptability problem of rough controller through researching the law of control rules. At last, the rough controller was used to control single inverted pendulum linear model system successfully.