本文選題：六維力傳感器　切入點：壓電式傳感器　出處：《光學精密工程》2017年05期 　論文類型：期刊論文

【摘要】：針對四點支撐結構的壓電式六維力傳感器線性度差,維間耦合嚴重的問題,提出了基于徑向基函數(shù)(RBF)神經(jīng)網(wǎng)絡的解耦算法。分析了耦合產(chǎn)生的主要原因,建立了RBF神經(jīng)網(wǎng)絡模型。通過對六維力傳感器進行標定實驗獲取解耦所需的實驗數(shù)據(jù),并對實驗數(shù)據(jù)進行處理。然后采用RBF神經(jīng)網(wǎng)絡優(yōu)化傳感器輸出系統(tǒng)的多維非線性解耦算法,解耦出傳感器的輸入輸出映射關系,得到解耦后的傳感器輸出數(shù)據(jù)。對傳感器解耦后的數(shù)據(jù)分析表明:采用RBF神經(jīng)網(wǎng)絡的解耦算法得到的最大Ⅰ類誤差和Ⅱ類誤差分別為1.29%、1.56%。結果顯示:采用RBF神經(jīng)網(wǎng)絡的解耦算法,能夠更加有效地減小傳感器的Ⅰ類誤差和Ⅱ類誤差,滿足了傳感器兩類誤差指標均低于2%的要求。該算法有效地提高了傳感器的測量精度,基本解決了傳感器解耦困難的難題,
[Abstract]:Aiming at the problem of poor linearity and serious coupling between dimensions of piezoelectric six-axis force sensors with four-point support structure, a decoupling algorithm based on radial basis function (RBF) neural network is proposed, and the main causes of coupling are analyzed. The RBF neural network model is established, and the experimental data for decoupling are obtained by calibrating the six-axis force sensor. Then the multi-dimensional nonlinear decoupling algorithm of sensor output system is optimized by using RBF neural network to decouple the input and output mapping relationship of the sensor. The output data of the sensor after decoupling are obtained. The analysis of the data after decoupling shows that the maximum class I error and the class 鈪，

本文編號：1557951