- 學位論文
主動式懸吊系統之灰預測自組織模糊滑動模式徑向基函數類神經網控制器的設計
Design of a Grey-Prediction Self-Organizing Fuzzy Sliding-Mode Radial Basis-Function Neural-Network Controller for Active Suspension Systems
摘要
自組織模糊控制器(self-organizing fuzzy controller, SOFC) 已被應用在控制工程領域上, 然而, 控制過程中SOFC 不易找尋適當的學習速率和權重因子。為了解決這個問題,本研究發展出灰預測自組織模糊滑動模式徑向基函數類神經網路控制器(grey-prediction self-organizing fuzzy sliding-mode radial basis-function neural-network controller, GPSFSRBNC)。該GPSFSRBNC 利用灰色預測演算法預測系統下一步的誤差做為控制器輸入。它不僅銷除了SOFC 和自組織模糊滑動模式控制器(self-organizing fuzzy slidingmodecontroller, SFSC) 因參數選擇不當引起的問題, 也解決了自組織模糊徑向基函數類神經網路控制器(self-organizing fuzzy radial basis-function neural-network controller, SFRBNC) 穩定性的問題。此外, 相較於自組織模糊滑動模式徑向基函數類神經網路(selforganizingfuzzy sliding-mode radial basis-function neural-network controller, SFSRBNC), GPSFSRBNC 因擁有預測的特性, 提供了更好的控制性能。GPSFSRBNC 已用於控制主 動式懸吊系統已確定其控制性能。經由模擬結果證實, GPSFSRBNC 比SOFC, SFSC, SFRBNC, SFSRBNC 以及被動式控制, 在乘坐汽車的舒適性及汽車的操控性上達到更好的控制性能。
並列摘要
Self-organizing fuzzy controllers (SOFCs) have been applied to the control engineering fields. However, it is difficult to find appropriate parameters of learning rate and weighting distribution for the design of an SOFC. To solve the problem, this study developed a grey prediction self-organizing fuzzy sliding-mode radial basis-function neuralnetwork controller (GPSFSRBNC). The GPSFRBNC uses a grey-prediction algorithm to predict the next step error of the system for the controller design. It not only eliminates the problem caused by the inappropriate selection of parameters in both an SOFC and a self-organizing fuzzy sliding-mode controller (SFSC), but also solves the stability problem of a self-organizing fuzzy radial basis-function neural-network controller (SFRBNC) application. Moreover, as compared with a self-organizing fuzzy sliding-mode radial basis-function neural-network (SFSRBNC), the GPSFRBNC has a predicted property, thereby providing better control performance. The GPSFRBNC was employed to control an active suspension system to determine its control performance. Simulation results demonstrated that the GPSFSRBNC achieved better control performance than the SOFC, SFSC, SFRBNC, SFSRBNC as well as passive control, in terms of the ride comfort and the road-holding capability of the vehicle for active suspension control.