CAJ | 학술논문

为改善门机系统的调速性能，提出一种与专家系统相结合的单神经元PI控制算法。该智能PI调节器设计简单，对被控对象模型依赖较小，通过在线边学习边控制的方式自动对PI参数优化，解决了传统固定参数PI控制的不足，实现了电梯门机永磁同步系统对不同转动惯量的适应性的控制要求。仿真试验结果表明，基于这种调节器的调速系统无论重门轻门，其响应速度较快、超调量较小、调节时间较短、具有较强的跟踪能力和良好的鲁棒性，其控制性能优于常规固定PI控制。
위개선문궤계통적조속성능，제출일충여전가계통상결합적단신경원PI공제산법。해지능PI조절기설계간단，대피공대상모형의뢰교소，통과재선변학습변공제적방식자동대PI삼수우화，해결료전통고정삼수PI공제적불족，실현료전제문궤영자동보계통대불동전동관량적괄응성적공제요구。방진시험결과표명，기우저충조절기적조속계통무론중문경문，기향응속도교쾌、초조량교소、조절시간교단、구유교강적근종능력화량호적로봉성，기공제성능우우상규고정PI공제。
According to the bad influence on system speed controlling performance which was caused by the sub-stantial change of rotational inertia in the different elevator door system, a control algorithm of the single neuron PI combined with expert systems was proposed. The design of the intelligent PI regulator is simple, and is less dependent on the plant model. It can optimize PI regulator parameter automatically, which overcomes the shortage of traditional PI controller, by means of online learning while controlling. The simulation results reveal that, the speed control sys-tem based on the intelligent PI can perform well on elevator doors of different weight. It has strong tracking capabili-ties and good robustness, and its control performance is superior to that of the conventional PI control.