Tension Robust Control Strategy Based on Self-optimizing Algorithm

In tension systems, the main concern is to decouple the tension and velocity in spite of radius variations and other perturbations. To solve this control problem and design the optimal controller, this paper develops a mixed-sensitivity robust H∞ control based on self-optimizing algorithm. First, the modeling of the tension system is presented with relative theorems. Second, a mixed-sensitivity robust H∞ control which reduces the coupling between tension and velocity is compared to PID controller. But the H∞ controller is conservative, the H∞ controller method with parameter fuzzification is introduced. At low velocity, the H∞ control method with parameter fuzzification gets fine results. However, for the high velocity and real time requirements of tension system, the H∞ control method with parameter fuzzification cannot keep the controller optimal always. Therefore the tension robust control strategy based on self-optimizing algorithm is proposed. The controller is optimized by hyper generation GA (HGGA) which reduces computing time. Furthermore, the current error and the change of the error are turned to compensate time delay. In order to test the effectiveness of the proposed algorithms, the tension experiment platform analog system is developed based on DSP (TMS320LF2407A) board. Finally the algorithms in this paper are tested in tension system platform, and the feasibility has been verified. Key-Words: tension system; robust control; parameter fuzzification; genetic algorithm (GA); hyper generation GA (HGGA); time delay; DSP;