CAJ | 학술논문

近空间用无刷直流电机（ BLDCM）受环境参数影响出现不确定性参数摄动和负载扰动，系统的控制性能降低。为消除不确定性因素的影响，提出了一种基于RBF网络补偿的自适应模糊控制算法。该控制算法是在自适应模糊控制的基础上，引入RBF网络补偿控制器，对参数摄动和负载转矩突变引起的转速误差进行在线辨识和动态补偿，以达到快速鲁棒自适应控制目的。对比具有RBF网络补偿的自适应模糊控制和自适应模糊控制的模拟仿真实验结果表明：在转速变化、负载转矩突变和转动惯量改变条件下，有RBF网络补偿控制的响应时间缩短了10 ms以上，响应过程中，电磁转矩的瞬时峰值减少了20％左右，对近空间BLDCM系统的不确定性鲁棒性强。
근공간용무쇄직류전궤（ BLDCM）수배경삼수영향출현불학정성삼수섭동화부재우동，계통적공제성능강저。위소제불학정성인소적영향，제출료일충기우RBF망락보상적자괄응모호공제산법。해공제산법시재자괄응모호공제적기출상，인입RBF망락보상공제기，대삼수섭동화부재전구돌변인기적전속오차진행재선변식화동태보상，이체도쾌속로봉자괄응공제목적。대비구유RBF망락보상적자괄응모호공제화자괄응모호공제적모의방진실험결과표명：재전속변화、부재전구돌변화전동관량개변조건하，유RBF망락보상공제적향응시간축단료10 ms이상，향응과정중，전자전구적순시봉치감소료20％좌우，대근공간BLDCM계통적불학정성로봉성강。
Because of the environmental parameters transformation, the parameters perturbation and load torque disturbances of the brushless direct current motor ( BLDCM) in near space will appear, and the response speed and stability of control system will be bad. To solve this problem, we propose an adaptive fuzzy control algorithm based on RBF( radial basis function) neural network compensation. The adaptive fuzzy controller is deduced to ensure the BLDCM system has good dynamic performance, the RBF neural network is adopted to do online identification and compensate for the speed error when the parameters perturbation and load torque disturbance appear in order to a-chieve the purposes of fast response speed and good robustness. Comparing the simulation results of adaptive fuzzy control with those of RBF neural network compensation and adaptive fuzzy control, we show preliminarily that:(1) the adaptive fuzzy control Based on RBF neural network has a strong robustness against the uncertainties of the BLDCM;(2) its response time is shorten by adaptive fuzzy control over 10ms;(3) its peak electromagnetic torque is decreased about 20% during the response process.