濾清器主要功能是將燃油系統中的雜質微粒濾除，長時間的使用下，濾清器中累積的雜質會造成濾清器嚴重的阻塞，進而影響油路系統中的其他元件。為避免此情況發生，需更換濾芯。過去都是以時間和里程判斷更換時機，本論文提出透過壓差量測達成油品汙染程度及濾芯阻塞程度之即時估測，主動提醒使用者油品狀況及更換濾芯的時機。
以擴展型卡爾曼濾波器(extended Kalman filter, EKF)為基礎所提出之估測器，因具有雜訊免疫及線上即時估測之能力，在已知文獻上廣泛被運用於狀態估測，但鮮少用於油品濾清器。濾清器模型以濾清器雜質累積程度和油品雜質濃度為估測狀態，流量為輸入，而濾清器上游和下游端的壓力差為量測輸出。模型中壓差來自三個部分，濾清器空殼管路、濾芯本體與雜質層。空殼與濾芯在固定流量下，壓差值能穩定在一定的誤差下，而雜質層的累積則會影響濾清器整體壓差變化。擴展型卡爾曼濾波器即利用雜質層造成的壓差變化來估測累積雜質的多寡及油品汙染程度，進而達到濾清器預警的作用。
實驗結果顯示，以擴展型卡爾曼濾波器為基礎之即時估測，可以有效的估算出濾清器雜質累積程度與油品汙染程度。

Oil filters are commonly used to remove the impurity particles in oil piping systems.
After a long time usage, the retained particles clog the filter medium and impact the elements in the oil piping system.
To avoid this situation, the filter needs to be replaced regularly, conventionally based on the usage period on the mileage.
In this thesis, a real time estimator is developed to estimate the oil contamination level and oil filter loading for filter replacement timing notification.
Extened Kalman filter (EKF) is widely used for estimation due to its noise imunity and real time estimation capability, but its application in the oil filter field is still rare.
The oil filter model consists of filter loading and contamination level as states, the flow rate as the input and the pressure drop across the filter as the measurement output.
The pressure drop is contributed by the piping in the filter case, the filter cartridge and the particle layer. Experimential results indicate that simultaneous estimation of the oil contamination level and the filter loading is achieved by the EKF-based estimator.

[1] 王衍凱, 以擴張型卡爾曼濾波器為基礎之感應馬達無感測控制及定子轉子阻抗估測,碩士論文,99年,
台北.
[2] 蔡瀛逸等, 柴油車氮氧化物、多環芳香烴化合物及粒狀污染物排放減量技術及策略之研究. 環保
署/國科會空汙防制科研合作計畫, 2012.
[3] 鄭偉德, 以數位訊號處理器實現以擴展型卡爾曼濾波器為基礎之超級電容SOC 與溫度動態即時估
測, 碩士論文,102年, 台北.
[4] 燃油濾清器先進驗證技術,2013.
[5] K.Ogata, “Discrete time control system,” Prentice Hall Inc..
[6] I. 19438:2003(E), “Developmentdiesel fuel and petrol filters for internal combustion engines
- filtration efficiency using particle counting and contaminant retention capacity,” INTER-
NATIONAL STANDARD.
[7] H. SAIDANI, Diesel fuel and petrol filters for internal combustion engines - Filtration effi-
ciency using particle counting and contaminant retention capacity, 2014.
[8] “The new generation of diesel engines,” May 2007, taiwan Maruboshi Information Technol-
ogy Co., Ltd.
[9] lmfilter, mK667920 Fuel Filter For MITSUBISHI,2016.
[10] TOTAL, http://www.buyoils.co.uk/ourshop/prod-2614885-TOTAL-AEROHYDRAULIC-
520-20-litre-pack.html.
[11] S. Bonne and N. Arnault, “Oil filter clogging rule - correlation between mileage and lab
test clogging,” SAE INTERNATIONAL.
[12] L. J. J. R. Z. Y. W. L. G. Jing, “Study on micro-pore filter blockage model for oil contam-
ination level testing under constant pressure,” IEEE Computer Society.
[13] K. J. B. a. J. H. J. Julian C. Tan, Cornelius N. Opris, “A study of the refeneration process
in diesel particulate traps using a copper fuel additive,” SAE INTERNATIONAL.
[14] R. Holdich, Fundamentals of Particle Technology, 2012.
[15] G.Welch and G.Bishop, “An introduction to the kalman filter,” Darpartment of computer
Science University of North Carolina at Chapel Hill.