發(fā)布時間：2018-02-21 15:13

  本文關(guān)鍵詞： 渦扇發(fā)動機 容錯方法 狀態(tài)空間模型 LQG/LTR控制器 性能恢復(fù)　出處：《南京航空航天大學(xué)》2016年碩士論文　論文類型：學(xué)位論文

【摘要】：本文針對某型渦扇航空發(fā)動機在氣路部件故障情況下,研究其在整個飛行包線內(nèi)的容錯控制方法。最終實現(xiàn)整個閉環(huán)控制系統(tǒng)在線的故障診斷和性能恢復(fù),在保證發(fā)動機穩(wěn)定安全工作的基礎(chǔ)上減少發(fā)動機推力以及喘振裕度等性能損失。基于容積動力學(xué)發(fā)動機部件級模型,研究建立狀態(tài)空間模型和提高線性化模型精度的方法。選取容腔中不同的熱力學(xué)參數(shù)作為狀態(tài)量建立狀態(tài)空間模型,并分析不同數(shù)量和不同類型的狀態(tài)量組合對于線性化精度的影響。在穩(wěn)態(tài)基點線性化模型的基礎(chǔ)上建立了慢車到最大狀態(tài)的大范圍小偏差模型,得到了較高的模型精度。針對航空發(fā)動機線性二次型高斯/回路傳遞恢復(fù)(LQG/LTR)多變量魯棒控制中需要手動選取設(shè)計參數(shù)的問題,提出基于量子行為粒子群優(yōu)化算法(QPSO)的LQG/LTR控制器設(shè)計方法。根據(jù)目標(biāo)回路與實際開環(huán)回路的奇異值曲線自動優(yōu)化可調(diào)參數(shù)q,獲得控制器各設(shè)計參數(shù)。該設(shè)計方法降低了傳統(tǒng)手動調(diào)試的盲目性,具有良好的控制性能�？紤]外物吸入、機械斷裂,或者發(fā)動機性能蛻化量達到一定程度時,可能引發(fā)的氣路部件故障,對易發(fā)生故障的高壓轉(zhuǎn)子部件根據(jù)效率和流量的下降程度劃分為高、中、低三個故障等級,提出離線設(shè)計性能恢復(fù)插值表并用于在線插值的方法對推力和喘振裕度進行恢復(fù)。對于高等故障采用QPSO算法計算發(fā)動機控制參數(shù)的調(diào)整值,對于中等及低等故障則利用近似函數(shù)求解,完成離線性能恢復(fù)插值表的設(shè)計。在此基礎(chǔ)上,設(shè)計基于擴展卡爾曼濾波(EKF)的故障診斷模塊,以及將控制調(diào)整量轉(zhuǎn)換為指令調(diào)整量的非線性轉(zhuǎn)換模塊。最終構(gòu)建發(fā)動機閉環(huán)容錯控制系統(tǒng),實現(xiàn)發(fā)動機高壓轉(zhuǎn)速系統(tǒng)故障情況下的在線性能恢復(fù),仿真表明該容錯控制可以有效減少故障后發(fā)動機推力以及喘振裕度的損失。
[Abstract]:In this paper, the fault-tolerant control method of a turbofan aero-engine in the whole flight envelope is studied under the condition of gas path failure. Finally, the on-line fault diagnosis and performance recovery of the whole closed-loop control system are realized. On the basis of ensuring the stable and safe operation of the engine, the performance loss of the engine such as thrust and surge margin is reduced. Based on the volumetric dynamics engine component level model, The methods of establishing the state space model and improving the precision of the linearization model are studied. Different thermodynamic parameters in the cavity are selected as the state variables to establish the state space model. The effects of different quantities and different types of state combinations on the linearization accuracy are analyzed. Based on the steady-state base point linearization model, a large range small deviation model from slow train to maximum state is established. In order to solve the problem that the design parameters need to be manually selected in the multivariable robust control of linear quadratic Gao Si / loop transfer recovery (LQG / LTR) of aeroengine, A design method of LQG/LTR controller based on quantum behavior particle swarm optimization algorithm (QPSO) is proposed. According to the singular value curve of the target loop and the actual open loop loop, the adjustable parameters Q are automatically optimized, and the design parameters of the controller are obtained. Reduces the blindness of traditional manual debugging, Has good control performance. Considering the suction of external matter, mechanical fracture, or engine performance degradation to a certain extent, may lead to gas circuit component failure, The components of high voltage rotor which are prone to fault are divided into three grades according to the decreasing degree of efficiency and flow rate: high, middle and low. An off-line design performance recovery interpolation table is proposed and used for on-line interpolation to recover thrust and surge margin. For higher faults, QPSO algorithm is used to calculate the adjustment value of engine control parameters. For medium and low faults, the off-line performance recovery interpolation table is designed by using approximate function. On this basis, a fault diagnosis module based on extended Kalman filter (EKF) is designed. Finally, the closed-loop fault-tolerant control system of the engine is constructed to realize the linear recovery in the case of the failure of the engine high-pressure speed system. Simulation results show that the fault tolerant control can effectively reduce the loss of thrust and surge margin after failure.
【學(xué)位授予單位】：南京航空航天大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2016
【分類號】：V263.6

【參考文獻】

相關(guān)期刊論文 前6條

1 王健康;張海波;孫健國;黃向華;;基于復(fù)合模型及FSQP算法的發(fā)動機性能尋優(yōu)控制試驗[J];推進技術(shù);2012年04期

2 張鵬;黃金泉;;航空發(fā)動機氣路故障診斷的平方根UKF方法研究[J];航空動力學(xué)報;2008年01期

3 袁春飛;姚華;劉源;;基于機載自適應(yīng)模型的航空發(fā)動機控制[J];推進技術(shù);2006年04期

4 李秋紅;孫健國;;基于遺傳算法的航空發(fā)動機狀態(tài)變量模型建立方法[J];航空動力學(xué)報;2006年02期

5 徐啟華,師軍;基于支持向量機的航空發(fā)動機故障診斷[J];航空動力學(xué)報;2005年02期

6 馮正平,孫健國;航空發(fā)動機小偏差狀態(tài)變量模型的建立方法[J];推進技術(shù);2001年01期

相關(guān)博士學(xué)位論文 前1條

1 趙晶;量子行為粒子群優(yōu)化算法及其應(yīng)用中的若干問題研究[D];江南大學(xué);2013年

相關(guān)碩士學(xué)位論文 前2條

1 王健康;基于機載復(fù)合模型及SQP的發(fā)動機性能尋優(yōu)控制研究[D];南京航空航天大學(xué);2010年

2 張華;LQG/LTR方法在渦扇發(fā)動機控制中的應(yīng)用研究[D];西北工業(yè)大學(xué);2004年

，

本文編號：1522198