| Hot-pressing is one of the most important processes in the middle density fiberboard manufacture industry. It is directly affected the quality and performance of fiberboard by designing and adjusting the parameters in hot-pressing process. Based on the principle of mass and heat transfer of fiberboard hot-pressing process, this paper was building the math model of heating stage and pressure of the Dry middle density fiberboard hot pressing process to built the controller and optimize the parameters in hot-pressing process.Because of the controlled object could maintain the characteristics of non-linearity and hysteresis, this paper was designing a fuzzy-PID controller and a fuzzy controller, which is used Mamdani inference synthesis algorithm, and Offline Programming method to simplify the fuzzy rules from reasoning synthesis operation to inquire the fuzzy control rule tables, respectively to control the pressure and the running time of the hot-pressing process. Comparing with the traditional PID controller, these control strategy could decrease the overshoot, improve the stability of the system, and it could also show the obvious superiority in improving the robustness of the system and adapt the fluctuation of the parameters. In terms of running time of the hot-pressing, the time-fuzzy controller could adjust the parameters around the set point with less fluctuation and could have stronger control quality than the tradition way which is controlled by digital signal.In this paper, fuzzy adaptive method was used to adjust the parameter of PID controller of pressure, and Dual Population Genetic Algorithm was used to optimize the fuzzy controller of the running time of the hot-pressing. The traditional PID parameter tuning method may have stronger subjectivity and blindness characteristic; however, fuzzy adaptive method, which has stronger ability to deal with the nonlinear of the system, could conquer these problems, which also includes the target of tune the PID parameters. In the aspect of timing fuzzy controller, by complex coding, Dual Population Genetic Algorithm could increase the amount of information carried by a single individual, which means the diversity of the population could be increased and the convergence time could be decreased. The simulation showed that the optimized fuzzy controller may have smaller overshoot, weaker nonlinear and hysteretic effect, at the same time, the dynamic performance, control quality and the robustness could be improved. |
