查詢結果分析
來源資料
頁籤選單縮合
題名 | Design of a Neural Fuzzy Controller for Dynamic Braking Resistor of Power Systems=設計一種類神經乏晰控制器作電力系統動態煞車電阻控制 |
---|---|
作者姓名(中文) | 王孟輝; | 書刊名 | 勤益學報 |
卷期 | 12 1994.11[民83.11] |
頁次 | 頁311-339 |
分類號 | 448.942 |
關鍵詞 | 類神經乏晰控制器; 動態煞車電阻; 暫態穩定度控制; Neural fuzzy controller; Dynamic braking resistor; Transient stability control; |
語文 | 英文(English) |
中文摘要 | 本論文設計一種結合類神經網路與乏晰理論的控制器,以改善電力系統 之暫態穩定度。首先,整個控制器的初始架構係根據發電機轉子角度和速度以乏 晰控制理論設計;其次,將乏晰控制法則轉換成一組輸出和輸入之間的訓練樣 本,再利用一種多層類神經網路及逆傳遞學習法學習想要的控制法則。基於類神 經網路平行和分散處理之計算能力,本文所提之類神經乏晰控制器能大量降低傳 統乏晰控制器冗長的推理時間;同時亦具有容忍局部故障之能力。最後,為證明 本文所提方法之實用性,以一單機電力系統為模擬對象,並和可變結構控制器比 較。模擬結果顯示本文所提之類神經乏晰控制器具強健性,並且其性能亦優於可 變結構控制器。 |
英文摘要 | A novel neural fuzzy controller (NFC) for dynamic braking resistor is presented to improve the transient stability of powersystems. The primitive control scheme is developed by the fuzzylogic theory according to the angle and speed on the generator.The fuzzy controller is of decentralized nature since only localmeasurements are employed as the input signals. Then the fuzzyrules expressed in linguistic variables are transformed into atraining set of input-output patterns. A multi-layer neural network (MNX) is applied to learn the desired control laws using theerror back propagation algorithm, the neural network technique isadvantageously used to refine and generalize the control rules offuzzy system. Besides, the rule-matching time of the inferenceengine in the traditional fuzzy logic system can be avoided. Todemonstrate the effectiveness of the proposed control scheme, acomparison to a variable structure controller (VSC) is analyzed.Numerical results show that the proposed neural fuzzy controllercan provide fairly good damping characteristic over wider ranges ofoperating conditions, and can yield better dynamic performance thanthe VSC. Significantly, with the evolving symbiosis of these newtechniques, the neural fuzzy controller has shown to be more adaptive and much less sensitive to changes in parameters. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。