【摘要】 电液驱动Stewart平台是一个多输入多输出复杂并联机构,各通道间存在严重负载交联耦合。在确定平台各通道性能参数时,按照简单的阀控缸系统进行理论计算或系统辨识,结果和实际情况相差极大,这严重影响平台控制系统的设计和系统性能的提高。为了解决上述问题,设计一种基于Levenberg-Marquardt寻优算法的六维互不相关信号激励的Stewart平台闭环辨识算法,并应用于电液驱动Stewart平台试验样机。试验结果表明该辨识算法准确地辨识出平台通道性能参数,取得较满意的辨识结果。更多还原

【Abstract】 Stewart platform driven by electro-hydraulic system is a multi-input and multi-output baroque parallel mechanism, in which serious load-coupling phenomenon occurs between channels. Generally performance parameters of each channel of platfrom are computed or identified according to the theory of servo-valve controlling cylinder system, the results obtained are different from practical operation conditions greatly, which affects the design and performance improvement of the platform control system severely. In order to solve the above problem, a kind of closed-loop identification algorithm of Stewart platform is designed on the basis of Levenberg-Marquardt optimization algorithm and a six-dimension uncorrelated exciting signals, and the experimental research on parameters identification is done on Stewart platform driven by electro-hydraulic test prototype. Experiment results show that the identification algorithm can identify the platform channels performance parameters precisely, and satisfactory identification results are obtained.更多还原