無人水下航行器執(zhí)行機(jī)構(gòu)故障診斷與容錯控制研究
發(fā)布時間：2018-11-26 13:33
【摘要】：海洋蘊藏著大量人類社會發(fā)展所需卻難以開采的自然資源,隨著科學(xué)技術(shù)的進(jìn)步,人類對海洋的探索與開發(fā)日新月著。水下航行器作為人類進(jìn)軍海洋的主要工具之一,受復(fù)雜海洋環(huán)境的影響,其運動控制系統(tǒng)中的執(zhí)行機(jī)構(gòu)可能發(fā)生各種故障,使得人員與設(shè)備的安全受到威脅。在面臨故障時水下航行器所采取的故障診斷和容錯控制措施,對提高人員和設(shè)備的安全保障具有重要意義。國內(nèi)外開展水下航行器的故障診斷與容錯控制研究已逾二十年,而較為系統(tǒng)解決執(zhí)行機(jī)構(gòu)故障問題的研究方案尚少。鑒于此,本文開展常見回轉(zhuǎn)體式和框體式無人水下航行器(UUV)執(zhí)行機(jī)構(gòu)的故障診斷與容錯控制研究,分別就UUV運動控制系統(tǒng)線性與非線性動力學(xué)模型的故障估計與檢測、回轉(zhuǎn)體式和框體式UUV運動控制系統(tǒng)執(zhí)行機(jī)構(gòu)的故障定位與辨識、執(zhí)行機(jī)構(gòu)發(fā)生故障但仍可控時的故障調(diào)節(jié)主動容錯控制以及故障執(zhí)行機(jī)構(gòu)存在冗余時的控制重構(gòu)主動容錯控制等問題進(jìn)行深入的研究,提出相應(yīng)的算法并進(jìn)行仿真實驗。本文的主要研究內(nèi)容如下：(1)研究模型控制輸入的故障估計與檢測問題。首先,從UUV運動控制系統(tǒng)的線性與非線性動力學(xué)模型著手,設(shè)計加性和乘性故障描述因子以表征模型控制輸入因執(zhí)行機(jī)構(gòu)故障而發(fā)生的改變；并先后采用H∞濾波估計算法、線性觀測估計算法和連續(xù)-離散無跡卡爾曼濾波估計算法,通過濾波器或者觀測器實時獲得線性動力學(xué)模型、狀態(tài)全部可知以及部分可知的非線性動力學(xué)模型的故障描述因子估計。然后,結(jié)合故障描述因子的估計,設(shè)計故障檢測函數(shù),實時對比檢測函數(shù)值與故障閾值實現(xiàn)故障檢測。最后,通過仿真實驗驗證模型控制輸入故障估計與檢測算法的有效性。(2)研究回轉(zhuǎn)體式UUV執(zhí)行機(jī)構(gòu)的故障定位與辨識問題。針對回轉(zhuǎn)體式UUV姿態(tài)控制與動力推進(jìn)兩個子系統(tǒng)的舵面、舵機(jī)、螺旋槳和推進(jìn)電機(jī)等機(jī)構(gòu),基于模型控制輸入的故障估計與檢測結(jié)果,結(jié)合傳感器量測數(shù)據(jù)分析控制輸入方程,在排除其它執(zhí)行機(jī)構(gòu)故障的前提下,定位出當(dāng)前執(zhí)行機(jī)構(gòu)的故障,并通過解控制輸入方程進(jìn)一步實現(xiàn)故障辨識。根據(jù)分析結(jié)果,采用仿真實驗驗證舵面形變故障定位方法的有效性。(3)研究框體式UUV執(zhí)行機(jī)構(gòu)的故障定位與辨識問題。針對框體式UUV通常配置的執(zhí)行機(jī)構(gòu)——推進(jìn)器,通過分析控制矩陣提出推進(jìn)器的冗余關(guān)系分析算法,并給出框體式UUV故障推進(jìn)器定位與辨識的前提；基于模型控制輸入的故障估計與檢測結(jié)果,分別研究僅配置標(biāo)量推進(jìn)器與配置有矢量推進(jìn)器的框體式UUV故障推進(jìn)器的定位與辨識問題；根據(jù)以上分析過程,歸納出框體式UUV故障推進(jìn)器的定位與辨識算法。分別采用僅配置標(biāo)量推進(jìn)器和配置有矢量推進(jìn)器的框體式UUV,建立仿真實驗,驗證所提故障推進(jìn)器定位與辨識算法的有效性。(4)基于執(zhí)行機(jī)構(gòu)的故障診斷研究,解決執(zhí)行機(jī)構(gòu)發(fā)生故障而UUV卻仍可控的容錯控制問題。針對執(zhí)行機(jī)構(gòu)發(fā)生故障卻仍然可控的情形,提出故障調(diào)節(jié)容錯控制算法,通過調(diào)節(jié)反饋控制輸入實現(xiàn)主動容錯控制,并應(yīng)用于舵面形變故障的主動容錯控制分析。針對故障執(zhí)行機(jī)構(gòu)存在配置冗余的情形,提出了無冗余約束的冗余補償與有冗余約束的冗余規(guī)劃容錯控制算法,前者采用冗余執(zhí)行機(jī)構(gòu)提供容錯控制輸入,后者從控制回路中刪除故障執(zhí)行機(jī)構(gòu)并重新規(guī)劃運動控制輸入。通過回轉(zhuǎn)體式UUV舵面形變故障和框體式UUV推進(jìn)器故障的容錯控制仿真實驗,分別驗證故障調(diào)節(jié)與控制重構(gòu)主動容錯控制算法的有效性。
[Abstract]:With the progress of science and technology, the exploration and development of the oceans and the oceans and the development of the oceans. As one of the main tools for human entering the sea, the underwater vehicle is affected by the complex marine environment, and the actuators in the motion control system may have various faults, so that the safety of personnel and equipment is threatened. The fault diagnosis and fault-tolerant control measures taken by the underwater vehicle in the face of failure are of great significance to the improvement of the safety and safety of personnel and equipment. The research on fault diagnosis and fault-tolerant control of underwater vehicle is over 20 years. In view of this, the fault diagnosis and fault-tolerant control of a common rotary body and a frame type unmanned underwater vehicle (UUV) actuator is carried out, and the fault estimation and detection of the linear and non-linear dynamic models of the UUV motion control system are respectively carried out. the fault location and identification of the rotary body and the actuator of the frame type UUV motion control system, the fault regulation active fault-tolerant control when the executing mechanism fails but is still controllable, and the control and reconstruction active fault-tolerant control when the fault executing mechanism is redundant, The corresponding algorithm is put forward and the simulation experiment is carried out. The main contents of this paper are as follows: (1) The problem of the estimation and detection of the control input of the model is studied. First, from the linear and non-linear dynamic model of the UUV motion control system, the additive and multiplicative fault description factors are designed to characterize the change of the model control input due to the failure of the actuator, and the H-frequency filtering estimation algorithm is adopted. The linear observation estimation algorithm and the continuous-discrete unscented Kalman filter estimation algorithm are used to obtain the linear dynamic model in real time by a filter or an observer, the state is all clear, and the fault description factor estimation of the partially-known nonlinear dynamic model is obtained. then, the fault detection function, the real-time comparison detection function value and the fault threshold value are combined to realize the fault detection in combination with the estimation of the fault description factor. and finally, the validity of the input fault estimation and detection algorithm is controlled by the simulation experiment verification model. (2) To study the fault location and identification of the rotary UUV actuator. aiming at the rudder surface, the steering engine, the propeller and the propulsion motor of the two subsystems of the rotary type UUV attitude control and the power propulsion, the input equation is controlled based on the fault estimation and detection result of the model control input, and the input equation is controlled in combination with the sensor measurement data analysis, Under the condition of excluding other actuator faults, the fault of the current executing mechanism is located, and the fault identification is further realized by the control of the input equation. Based on the results of the analysis, the effectiveness of the rudder surface deformation fault location method is verified by means of simulation. (3) To study the fault location and identification of the frame UUV actuator. According to the actuator _ propeller which is usually configured for the frame type UUV, the redundant relation analysis algorithm of the propeller is put forward by the analysis control matrix, and the premise of the positioning and identification of the frame type UUV fault propeller is provided; and the fault estimation and detection result input by the model control is carried out based on the model control input, In this paper, the positioning and identification of a frame type UUV (UUV) fault propeller with a vector propulsor are respectively configured. According to the above analysis process, the positioning and identification algorithm of the frame type UUV fault propeller is summarized. In this paper, only a scalar propeller and a frame type UUV equipped with a vector propeller are used to establish a simulation experiment to verify the effectiveness of the proposed positioning and identification algorithm. (4) Based on the fault diagnosis of the actuator, the fault-tolerant control problem of UUV is still controlled by the failure of the actuator. The fault-tolerant control algorithm is proposed to control the fault of the actuator, and the active fault-tolerant control is realized by adjusting the feedback control input, and the active fault-tolerant control analysis is applied to the deformation fault of the rudder surface. Aiming at the situation that the fault executing mechanism has the configuration redundancy, the redundant compensation and the redundant planning fault-tolerant control algorithm with redundant constraint are put forward, the former adopts the redundant executing mechanism to provide the fault-tolerant control input, The latter deletes the failed actuator from the control loop and reprogram the motion control input. The validity of the fault regulation and control reconfiguration active fault-tolerant control algorithm is verified by the fault-tolerant control simulation experiment of the rotary type UUV rudder surface deformation fault and the frame type UUV thruster fault.
【學(xué)位授予單位】：西北工業(yè)大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2015
【分類號】：U672
[Abstract]:With the progress of science and technology, the exploration and development of the oceans and the oceans and the development of the oceans. As one of the main tools for human entering the sea, the underwater vehicle is affected by the complex marine environment, and the actuators in the motion control system may have various faults, so that the safety of personnel and equipment is threatened. The fault diagnosis and fault-tolerant control measures taken by the underwater vehicle in the face of failure are of great significance to the improvement of the safety and safety of personnel and equipment. The research on fault diagnosis and fault-tolerant control of underwater vehicle is over 20 years. In view of this, the fault diagnosis and fault-tolerant control of a common rotary body and a frame type unmanned underwater vehicle (UUV) actuator is carried out, and the fault estimation and detection of the linear and non-linear dynamic models of the UUV motion control system are respectively carried out. the fault location and identification of the rotary body and the actuator of the frame type UUV motion control system, the fault regulation active fault-tolerant control when the executing mechanism fails but is still controllable, and the control and reconstruction active fault-tolerant control when the fault executing mechanism is redundant, The corresponding algorithm is put forward and the simulation experiment is carried out. The main contents of this paper are as follows: (1) The problem of the estimation and detection of the control input of the model is studied. First, from the linear and non-linear dynamic model of the UUV motion control system, the additive and multiplicative fault description factors are designed to characterize the change of the model control input due to the failure of the actuator, and the H-frequency filtering estimation algorithm is adopted. The linear observation estimation algorithm and the continuous-discrete unscented Kalman filter estimation algorithm are used to obtain the linear dynamic model in real time by a filter or an observer, the state is all clear, and the fault description factor estimation of the partially-known nonlinear dynamic model is obtained. then, the fault detection function, the real-time comparison detection function value and the fault threshold value are combined to realize the fault detection in combination with the estimation of the fault description factor. and finally, the validity of the input fault estimation and detection algorithm is controlled by the simulation experiment verification model. (2) To study the fault location and identification of the rotary UUV actuator. aiming at the rudder surface, the steering engine, the propeller and the propulsion motor of the two subsystems of the rotary type UUV attitude control and the power propulsion, the input equation is controlled based on the fault estimation and detection result of the model control input, and the input equation is controlled in combination with the sensor measurement data analysis, Under the condition of excluding other actuator faults, the fault of the current executing mechanism is located, and the fault identification is further realized by the control of the input equation. Based on the results of the analysis, the effectiveness of the rudder surface deformation fault location method is verified by means of simulation. (3) To study the fault location and identification of the frame UUV actuator. According to the actuator _ propeller which is usually configured for the frame type UUV, the redundant relation analysis algorithm of the propeller is put forward by the analysis control matrix, and the premise of the positioning and identification of the frame type UUV fault propeller is provided; and the fault estimation and detection result input by the model control is carried out based on the model control input, In this paper, the positioning and identification of a frame type UUV (UUV) fault propeller with a vector propulsor are respectively configured. According to the above analysis process, the positioning and identification algorithm of the frame type UUV fault propeller is summarized. In this paper, only a scalar propeller and a frame type UUV equipped with a vector propeller are used to establish a simulation experiment to verify the effectiveness of the proposed positioning and identification algorithm. (4) Based on the fault diagnosis of the actuator, the fault-tolerant control problem of UUV is still controlled by the failure of the actuator. The fault-tolerant control algorithm is proposed to control the fault of the actuator, and the active fault-tolerant control is realized by adjusting the feedback control input, and the active fault-tolerant control analysis is applied to the deformation fault of the rudder surface. Aiming at the situation that the fault executing mechanism has the configuration redundancy, the redundant compensation and the redundant planning fault-tolerant control algorithm with redundant constraint are put forward, the former adopts the redundant executing mechanism to provide the fault-tolerant control input, The latter deletes the failed actuator from the control loop and reprogram the motion control input. The validity of the fault regulation and control reconfiguration active fault-tolerant control algorithm is verified by the fault-tolerant control simulation experiment of the rotary type UUV rudder surface deformation fault and the frame type UUV thruster fault.
【學(xué)位授予單位】：西北工業(yè)大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2015
【分類號】：U672
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 張君昌,侯義斌,李言俊;基于知識熵漲落的系統(tǒng)智能容錯控制法[J];彈箭與制導(dǎo)學(xué)報;2000年02期
2 黃孝彬,牛玉廣,孫萬云;容錯控制及其在火電廠中的應(yīng)用[J];電力情報;2000年04期
3 王峰;汪t，
本文編號：2358700
本文鏈接：http://sikaile.net/kejilunwen/chuanbolw/2358700.html
最近更新
教材專著
