A new digital approach to design multivariable robust optimal control systems

This paper presents a new design of robust optimal controller for multivariable system. The row characteristic functions of a linear multivariable system and dynamic decoupling of its equivalent system, were applied to change the transfer function matrix of a closed-loop system into a normal function matrix, so that robust H^∞ optimal stability is guaranteed. Furthermore, for the decoupled equivalent control system the I^∞ optimization approach is used to have the closed-loop system embody optimal time domain indexes. A successful application on a heater control system verified the excellence of the new control scheme.

溶出曲线的常见评价方法及优劣性探讨

目的:对溶出曲线常见评价方法的性能进行比较分析,并探讨当溶出数据呈较高变异时,无法使用f_(2)因子的情况下,哪种替代方法较为合适。方法:以双环醇片作为实例,收集20批溶出数据,使用不同评价方式进行比较,包括非模型依赖方式、模型依赖差异性比较方式和等效性比较方式(建立模型参数等效区间)。结果:在溶出数据变异较大(>10%)时,非模型依赖法中的f_(2)-bootstrap法的结果与多变量统计距相比,更接近f_(2)因子法的结果;模型依赖法比较中的t检验法和多变量统计矩法的结果,与f_(2)因子法有较大差异;首次利用模型参数获得的各参数容忍区间;可作为模型参数的等效区间供今后溶出一致性的评估。结论:当选择非模型依赖法时,当原f_(2)因子法不适用时,f_(2)-bootstrap法可作为f;因子法的替代方法;新提出模型参数的等效区间能够弥补t检验的不足,更好地从质量等效层面上评价溶出曲线。