本文關鍵詞：氣象雷達散熱系統(tǒng)動力裝置的性能研究　出處：《中國民航大學》2016年碩士論文　論文類型：學位論文

  更多相關文章： 電機 故障預測 粒子群 支持向量機

【摘要】：隨著航空事業(yè)的發(fā)展,航空安全問題受到廣泛關注,而機載設備的安全性和可靠性直接影響航空安全。機載氣象雷達是機載航空電子系統(tǒng)的重要子系統(tǒng),雷達的損壞極有可能導致飛機中途返航,影響航班行程,給航空公司和機場帶來巨大損失,所以預測故障的發(fā)生時間以及趨勢,提前對失效航空器件進行維修、替換顯得尤為重要。氣象雷達系統(tǒng)失效的主要成因是散熱系統(tǒng)故障,AMETEK航空風扇作為散熱系統(tǒng)的核心部件,由永久分相式電容電機驅動,電機運行狀態(tài)異常成為影響氣象雷達工作效能的重要原因。因此快速有效地對電容式電機進行故障預測具有現(xiàn)實意義。本文在研究散熱系統(tǒng)結構特征的基礎上,通過分析氣象雷達散熱系統(tǒng)和其動力裝置電容式電機的故障機理,采用統(tǒng)計過程控制方法對電容式電機的性能進行預判,提醒維修人員對氣象雷達散熱系統(tǒng)進行實時監(jiān)測。同時運用改進的粒子群優(yōu)化算法模型,優(yōu)化支持向量機的懲罰因子和核函數(shù),提高支持向量機回歸預測精度,再進一步將支持向量回歸機和支持向量分類機相結合,對電容式電機的電壓、電流數(shù)據(jù)進行回歸預測,建立電機故障預測模型,從而實現(xiàn)電容式電機運行的狀態(tài)預測。通過實驗仿真和現(xiàn)場驗證,結果表明,基于支持向量機的電機故障預測模型可對氣象雷達散熱系統(tǒng)動力裝置電機的故障作出快速準確預測,以便保障氣象雷達系統(tǒng)正常工作,對提高飛行安全有較好的實用價值。
[Abstract]:With the development of aviation industry, aviation safety has been paid more and more attention, and the safety and reliability of airborne equipment have a direct impact on aviation safety. Airborne weather radar is an important subsystem of airborne avionics system. Radar damage is likely to lead to the mid-way return of aircraft, affect the flight itinerary, and bring huge losses to airlines and airports, so predict the time and trend of the failure, and repair the failed aviation devices ahead of time. The main cause of the failure of meteorological radar system is that the AMETEK aeronautical fan, as the core part of the heat dissipation system, is driven by permanent phase separation capacitor motor. The abnormal operating state of the motor has become an important reason to affect the operational efficiency of meteorological radar. Therefore, it is of practical significance to predict the fault of capacitive motor quickly and effectively. This paper studies the structural characteristics of the heat dissipation system. By analyzing the fault mechanism of capacitive motor of meteorological radar heat dissipation system and its power device, the performance of capacitive motor is forecasted by statistical process control method. At the same time, the improved particle swarm optimization algorithm model is used to optimize the penalty factor and kernel function of support vector machine to improve the prediction accuracy of support vector machine regression. Furthermore, the support vector regression machine and the support vector classifier are combined to predict the voltage and current data of the capacitive motor, and the fault prediction model of the motor is established. In order to realize the state prediction of capacitive motor operation, the experimental simulation and field verification show that. The motor fault prediction model based on support vector machine (SVM) can predict the motor fault of the heat dissipation system of meteorological radar quickly and accurately, so as to ensure the normal operation of meteorological radar system. It has good practical value to improve flight safety.
【學位授予單位】：中國民航大學
【學位級別】：碩士
【學位授予年份】：2016
【分類號】：TN959.4
，

本文編號：1359661