控制理论与应用
控製理論與應用
공제이론여응용
CONTROL THEORY & APPLICATIONS
2009年
12期
1378-1382
,共5页
董领逊%窦丽华%陈杰%夏元清
董領遜%竇麗華%陳傑%夏元清
동령손%두려화%진걸%하원청
混杂模型预测控制%间隙模式%PWA模型%动态规划
混雜模型預測控製%間隙模式%PWA模型%動態規劃
혼잡모형예측공제%간극모식%PWA모형%동태규화
hybrid model predictive control%backlash mode%PWA model%dynamic programming
研究含间隙机械系统的混杂模型预测控制问题.首先,将含间隙机械系统的运行模式分为"间隙模式"和"接触模式".其次,建立了含间隙机械系统的混杂分段仿射 (PWA)模型.然后,利用模型预测控制 (MPC)的方法对约束PWA系统的最优控制进行求解,通过动态规划与多参数二次规划方法,得到了MPC的离线解.最后,通过将分段二次 (PWQ)Lyapunov函数的求解转换成半正定规划,找到了确保闭环控制稳定性的PWQ Lyaplanov函数.跟踪参考速度的实验结果表明,混杂模型预测控制器对含间隙机械系统的跟踪控制具有较好的效果,能够满足小采样时间系统的实时控制要求.
研究含間隙機械繫統的混雜模型預測控製問題.首先,將含間隙機械繫統的運行模式分為"間隙模式"和"接觸模式".其次,建立瞭含間隙機械繫統的混雜分段倣射 (PWA)模型.然後,利用模型預測控製 (MPC)的方法對約束PWA繫統的最優控製進行求解,通過動態規劃與多參數二次規劃方法,得到瞭MPC的離線解.最後,通過將分段二次 (PWQ)Lyapunov函數的求解轉換成半正定規劃,找到瞭確保閉環控製穩定性的PWQ Lyaplanov函數.跟蹤參攷速度的實驗結果錶明,混雜模型預測控製器對含間隙機械繫統的跟蹤控製具有較好的效果,能夠滿足小採樣時間繫統的實時控製要求.
연구함간극궤계계통적혼잡모형예측공제문제.수선,장함간극궤계계통적운행모식분위"간극모식"화"접촉모식".기차,건립료함간극궤계계통적혼잡분단방사 (PWA)모형.연후,이용모형예측공제 (MPC)적방법대약속PWA계통적최우공제진행구해,통과동태규화여다삼수이차규화방법,득도료MPC적리선해.최후,통과장분단이차 (PWQ)Lyapunov함수적구해전환성반정정규화,조도료학보폐배공제은정성적PWQ Lyaplanov함수.근종삼고속도적실험결과표명,혼잡모형예측공제기대함간극궤계계통적근종공제구유교호적효과,능구만족소채양시간계통적실시공제요구.
The hybrid model-predictive-control for mechanical system with backlash is studied. The operation model of a mechanical system with backlash is distinguished into the "backlash mode" and the "contact mode", and a piecewise affine (PWA) model of the mechanical system with backlash is built. For controlling this constrained PWA system, the optimal control law is derived by using the model predictive control (MPC) method; the offline solution of which is obtained by using the dynamic programming and the multi-parametric quadratic programming method. Finally, a piecewise quadratic (PWQ) Lyapuno function for the stability of the closed-loop system is found by transforming its searching process to a semi-definite programming. Experiments in tracking a reference speed signal demonstrate that the hybrid MFC gives desirable performances for the mechanical system with backlash, satisfying the real-time control requirements for systems with small sampling time.